Does folic acid before pregnancy prevent birth defects?

Yes, and this is the strongest single result in the whole of nutrition. Taking folic acid before conception cuts neural tube defects, the malformations of the spine and brain that include spina bifida, to less than half. The trials and the observational studies agree on the number, which almost never happens, and dozens of countries now put folic acid into flour because of it.

Established. Pooling ten systematic reviews covering 296,816 women, folic acid or a multivitamin before conception gave a pooled effect of 0.43. Folic acid on its own came out at 0.23. The reviews built from randomised trials gave 0.43 and the reviews built from cohorts gave 0.45.

The part that gets missed: when

The neural tube closes in the third and fourth week after conception, which for most people is before a pregnancy test turns positive and often before a period is missed. A supplement started at the first antenatal appointment is too late for this particular job.

That is why the recommendation is written the way it is: 400 micrograms a day of folic acid from a supplement or fortified food, for everyone who could become pregnant, whether or not they are planning to. Roughly half of pregnancies are not planned, and the biology does not wait for the plan.

Why the synthetic form, and not just food

This is the one place on this site where the research points at the synthetic version rather than at the food. About 85 per cent of folic acid is absorbed when taken with food, against roughly 50 per cent of the folate that occurs naturally in leaves and beans.

The prevention trials used folic acid, not spinach. That does not make food folate useless, and it does mean the recommendation is deliberately written in the form that was tested.

The gap in that comparison is smaller than the label says

Folate is the one nutrient whose unit is a calculation rather than a measurement. A label does not count folate, it converts it. One microgram of folic acid is written down as 1.7 micrograms of food folate, a formula the Institute of Medicine set in 1998 and the FDA adopted because a plain microgram ignored the difference in how well the two are absorbed. Read backwards, that factor of 1.7 assumes folate from food is worth about 59 per cent of folate from a pill.

Unsettled. Measured directly, food folate does better than that. Seventy-two adults ate either a high-folate diet or a low-folate one topped up with folic acid for four weeks, with labelled folic acid given to everyone as a tracer. Folate from fruit, vegetables and liver came out at 78 per cent of folic acid by the labelled measure and 85 per cent by the change in serum folate. The authors concluded that folate-rich food lifts a population's folate status more efficiently than is generally assumed. Two confidence intervals that wide cannot rewrite the label on their own.

There is also no single number that would fit every food. In a crossover study, 24 adults each ate spinach, wheat germ and camembert with the folate tagged so it could be followed into the blood. Spinach came out ahead on both the amount absorbed and the speed, wheat germ next, camembert last, and the variation between people was moderate for spinach and very high for wheat germ. A fixed conversion is a working assumption rather than a property of food.

None of that changes the advice at the top of this page. The trials that prevented neural tube defects used folic acid, and a supplement is still what the recommendation is written in. What it changes is the tone in which the gap should be described: food folate is weaker than the pill, and not nearly as weak as the arithmetic behind a label implies.

Established. Population-level evidence backs it. Since mandatory folic acid fortification began in the United States in 1998, neural tube defect rates fell by 28 per cent. That is a whole country's worth of data pointing the same direction as the trials.

What folic acid does not do

The success of the neural tube result has pulled a lot of other claims along behind it, and they are weaker.

Unsettled. On the child's brain, 39 observational studies found better intellectual development and less autism, ADHD and language trouble in the children of mothers who took folic acid. Those are cohorts, not trials, and the same analysis found that taking far more than recommended added nothing and might do harm. So this is a reason to take the recommended dose rather than a reason to take more.

Myth. On bowel cancer, the story is often told as though folic acid prevents it. Eating more total folate does track with less colorectal cancer, at a relative risk of 0.84 in the general population. But in people who have already had a polyp removed, folic acid supplements did not reduce recurrence, with a relative risk of 1.05 on high-quality evidence. Prevention in healthy people and treatment after a polyp are different questions with different answers.

Where the research went next

The neural tube answer has been settled for thirty years, and settled questions stop attracting studies. So everything published on folic acid in pregnancy lately is about something else, and all of it is weaker. This section is here to keep the two apart, because a new paper is easy to mistake for an update to the old one.

Would a bigger dose do more?

Unsettled. Three trials gave women trying to conceive 4 to 5 mg of folic acid a day instead of the standard 0.4 mg. Among the 1,488 who went on to conceive, miscarriage was about a quarter less common, a risk ratio of 0.73, and the chance of becoming pregnant at all was 10 per cent higher. Counting everyone who had been randomised instead of only the women who conceived, which is the analysis that avoids choosing the group after the fact, the miscarriage figure moved to 0.79 and its interval crossed no effect. Three trials, low to moderate certainty, and a result that depends on which count you accept.

That last detail is the whole story of this finding, so it is worth saying in plain words. Comparing only the women who became pregnant means comparing two groups that were sorted partly by the treatment being studied. The cleaner count includes everyone from the start, and when it was done that way the miscarriage benefit faded. The pregnancy rate held up in both.

The one high-dose trial with a worrying number in it

A placebo-controlled trial at 70 obstetric sites gave 4 to 5.1 mg a day from early pregnancy, and 462 of the women were carrying twins. Pre-eclampsia in that group came out at 17.2 per cent on folic acid against 9.9 per cent on placebo, a crude relative risk of 1.75. Once the analysis was adjusted the figure fell to 1.58 and stopped being significant, and the authors state that the suggested raised risk cannot be confirmed.

This was a subgroup of a larger trial, so it was never built to answer the question. Treat it as the absence of any benefit, plus a number nobody can explain away. What it rules out is the idea that a large dose is a free bet.

What most prenatal trials now compare

Recent research on prenatal supplements mostly compares packages. The control arm is iron plus folic acid, and the other arm is a multivitamin containing both of those and a dozen more nutrients. That design measures what the extra micronutrients add on top of folic acid, and it cannot measure what folic acid itself does, because everybody in it got folic acid.

Established. Against iron and folic acid alone, the fuller supplement cut the most dangerous category of newborn, preterm and small for gestational age, by 29 per cent across 14 trials and 42,618 mother-infant pairs. Babies were 0.05 cm longer at birth, held a small length advantage to six months and then lost it, and stunting at three months was 14 per cent less common. Every one of those trials ran in a low- or middle-income country.

Two cautions come with that. The differences in growth are small in absolute terms and gone after six months, and the setting was places where deficiency is common, which is where supplements generally do their best work. None of it transfers to a well-fed woman choosing between two boxes in a pharmacy.

The one situation where folic acid makes a treatment fail

Not for everyone. Where malaria is treated with antifolate drugs, most often sulfadoxine-pyrimethamine, folic acid taken alongside makes the treatment fail more often. Pooling eight trials in 3,486 people, treatment failure was about twice as likely at day 7 and 35 per cent more likely at day 28, and parasites cleared more slowly at day 3. All four outcomes were rated moderate certainty. Most of the trials used more than the 1,000 microgram upper limit, so this is a caution about high-dose folic acid during malaria treatment. Anyone in that position should tell the clinician treating the malaria what they take.

Methylfolate, MTHFR and the thing people are worried about

The prenatal aisle now sells a second form, 6S-5-methylfolate, usually at a higher price and usually alongside a claim about unmetabolised folic acid. Unmetabolised folic acid is what it sounds like: folic acid circulating in the blood that the body has not yet converted. The question worth answering is whether it does anything, and then whether swapping forms is worth doing.

Established. Swapping does lower it. Three trials comparing equal amounts of the two forms found less unmetabolised folic acid in the blood on methylfolate, with a standardised difference of 0.99, rated low certainty by GRADE. A 24-week trial in pregnancy found the same thing: detectable unmetabolised folic acid in 7 per cent of mothers on the methylfolate prenatal against 31 per cent on the folic acid one, and lower concentrations in the placenta.

Unsettled. Whether that matters is the part nobody has settled. In the same pregnancy trial, total folate and 5-methylfolate in the mother's serum, in the cord blood and in the placenta at week 24 came out no different between the two forms. The supplement did the narrow thing it is sold to do and nothing measurable beyond it.

The case against unmetabolised folic acid rests on observational studies, and there are three reasons they cannot carry it. Folic acid raises unmetabolised folic acid and total folate together, so the two cannot be separated in data where nobody assigned the dose. Unmetabolised folic acid is at most 1 to 2 per cent of the folate in blood, which makes measurement error large against its range, and a single sample reflects when the last tablet was taken as much as how big it was. In the cohort most often quoted, the timing was never recorded.

Myth. High folic acid intake leaves a harmful residue in the blood that observational studies have detected. The systematic review that classified this literature by exposure found the associations inconsistent, confounded by total folate and never replicated. That is a reason to keep studying it. It is not a finding.

Unsettled. MTHFR genotype testing sits in the same place, and the review says so in as many words: it calls the genotype question a research question. People with the TT variant, roughly one in ten of European ancestry and more in Mexican and Mediterranean populations, may show a larger homocysteine response to folic acid. The pattern did not hold inside every trial and no pooled genotype trend was estimated. A genotype result does not yet pick your supplement.

Put together: if you prefer methylfolate, the trial evidence says you will get the same folate into the same places with less unmetabolised folic acid, and nobody can tell you that buys you anything. If you take folic acid, the reason to stay near 400 micrograms is the B12 problem below, which is a documented harm rather than a suspected one.

The caution that matters most

Not for everyone. Large amounts of folic acid can correct the anaemia of vitamin B12 deficiency while leaving the nerve damage to progress unnoticed. That is the reason the 1,000 microgram upper limit exists, and it applies to the synthetic form only. It matters most for older adults, for people who eat no animal products and for people on long-term metformin or acid-suppressing drugs, all of whom are the people most likely to be low in B12 in the first place.

The reason that trap exists is that the two deficiencies look identical on a blood count. The main sign of folate deficiency is megaloblastic anaemia, where red blood cells are large and abnormal, and that is exactly what B12 deficiency produces too. Folic acid fixes the picture in the blood and does nothing for the nerves.

Low B12 combined with high folic acid has been associated with roughly two to three times higher risk of cognitive impairment in older adults, which is an observational finding and a good reason not to take large folic acid doses without checking B12 first.

What to do with this

If you could become pregnant

Four hundred micrograms a day of folic acid from a supplement or fortified food, starting before you are trying rather than after. Food folate on top of that is good for you and is not what the recommendation is written in.

If you have had a pregnancy affected by a neural tube defect

The dose is different and much higher, and it comes from your doctor rather than from a shelf. The pooled evidence for preventing a recurrence sits on smaller numbers than the evidence for preventing a first occurrence.

If you are not planning a pregnancy at all

Folate from food is enough. The 400 microgram recommendation is written for a specific job, and the upper limit of 1,000 micrograms for the synthetic form is worth respecting, particularly if your B12 status is uncertain.

If you drink heavily or take methotrexate

Alcohol interferes with folate absorption, speeds its breakdown and increases losses in urine. And people taking methotrexate for cancer should speak to their oncologist before any folate supplement, which is a different situation from the low-dose folic acid given deliberately alongside methotrexate for rheumatoid arthritis.

What would change this answer

• Nothing is likely to overturn the neural tube result. Fortification has made new randomised trials both unethical and impractical in countries that fortify, which is what success looks like.
• The neurodevelopment claims need randomised evidence, and the observational signal is strong enough that somebody should try. Until then, the recommended dose is where the evidence is.
• The upper end deserves better data. Several findings point at high folic acid intakes being unhelpful or harmful, including in the neurodevelopment analysis, and none of them come from trials designed to test it.

How we searched

Sources

1. fol-t1 · Meta-analysis or systematic review · umbrella review of systematic reviews and meta-analyses · n = 296,816
   The random-effects model analysis from 10 included systematic review and meta-analysis revealed that the pooled effect of either folic acid or multivitamin supplementation for the prevention of NTD globally is found to be 0.43 (95% CI: 0.29, 0.58) (I2 = 93.50%; p ≤0.001).
   
   Who: women taking folic acid or multivitamins before and around conception, 10 reviews
   Effect: pooled effect 0.43 (95% CI 0.29 to 0.58) for folic acid or multivitamins; 0.23 (0.09 to 0.37) for folic acid alone; heterogeneity 93.5 per cent
   Certainty: umbrella review of reviews, high heterogeneity, AMSTAR used for quality
   Tesfa et al., Neuroepidemiology, 2025 · checked 2026-09-16 · we read the abstract
2. fol-t2 · Meta-analysis or systematic review · umbrella review, subgroup analysis by study design · n = 296,816
   The pooled effect of either folic acid or multivitamin supplementation for the prevention of NTD was found to be 0.45 (0.03, 0.87) in SRMs of observational studies, while this estimate is 0.43 (0.32, 0.54) among SRMs of randomized controlled
   
   Who: reviews of randomised trials and of observational studies
   Effect: pooled effect 0.43 (95% CI 0.32 to 0.54) in reviews of randomised trials and 0.45 (0.03 to 0.87) in reviews of observational studies
   Certainty: agreement between trial-based and observational reviews
   Tesfa et al., Neuroepidemiology, 2025 · checked 2026-09-16 · we read the abstract
3. fol-t3 · Observational data · systematic review and meta-analysis of cohort and case-control studies · n = 39 studies
   We also found that FA over-supplementation was not associated with an improvement in offspring's brain development, and may have a negative impact on offspring's neurodevelopmental outcomes.
   
   Who: mothers and their children, 32 cohort studies and 7 case-control studies
   Effect: improved intellectual development and reduced risk of autistic traits, ADHD, behavioural and language problems; over-supplementation not associated with improvement and possibly harmful
   Certainty: observational data only, no randomised evidence
   Liu et al., Critical Reviews in Food Science and Nutrition, 2023 · checked 2026-09-16 · we read the abstract
4. fol-t4 · Observational data · umbrella review of meta-analyses of observational studies and randomised trials · n = 10 associations
   In the general population, moderate-quality evidence supported an inverse association between total folate intake (from foods and supplements) and CRC risk (RR 0.84; 95% CI 0.80 to 0.90)
   
   Who: general population and people with inflammatory bowel disease or a history of adenoma, 5 meta-analyses
   Effect: total folate intake and colorectal cancer relative risk 0.84 (95% CI 0.80 to 0.90), moderate quality; adenoma recurrence with folic acid supplements 1.05 (0.86 to 1.29), high GRADE rating
   Certainty: moderate for the dietary association, high for the null result on adenoma recurrence
   Wang et al., BMJ Open, 2025 · checked 2026-09-16 · we read the abstract
5. fol-01 · Position of an expert body
   Folate functions as a coenzyme or cosubstrate in single-carbon transfers in the synthesis of nucleic acids (DNA and RNA) and metabolism of amino acids.
   
   Who: general
   Effect: physiological role
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
6. fol-02 · Position of an expert body
   400 mcg DFE for males and females age 19+; 600 mcg DFE/day during pregnancy
   
   Who: adults and pregnancy
   Effect: 400 / 600 mcg DFE per day
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
7. fol-03 · Position of an expert body
   The FNB did not establish a UL for folate from food because high intakes of folate from food sources have not been reported to cause adverse effects.
   
   Who: adults
   Effect: UL 1,000 mcg folic acid
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
8. fol-04 · Position of an expert body
   At least 85% of folic acid is estimated to be bioavailable when taken with food, whereas only about 50% of folate naturally present in food is bioavailable.
   
   Who: general
   Effect: 85% vs 50% bioavailability
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
9. fol-05 · Randomised controlled trial(s)
   clinical trial evidence shows that adequate periconceptional folic acid consumption by women prevents a substantial proportion of NTDs.
   
   Who: women capable of becoming pregnant
   Effect: prevention of NTDs
   NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
10. fol-06 · Position of an expert body
    consume 400 mcg of folic acid daily from supplements, fortified foods, or both
    
    Who: women capable of becoming pregnant
    Effect: 400 mcg/day folic acid advised
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
11. fol-07 · Observational data
    Since 1998, when mandatory folic acid fortification began in the United States, NTD rates have declined by 28%.
    
    Who: US population
    Effect: −28% NTD rate
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
12. fol-08 · Position of an expert body
    Large amounts of folate can correct the megaloblastic anemia, but not the neurological damage, that can result from vitamin B12 deficiency.
    
    Who: people with B12 deficiency
    Effect: masking of B12 deficiency
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
13. fol-09 · Position of an expert body
    alcohol interferes with folate absorption and hepatic uptake, accelerates folate breakdown, and increases its renal excretion
    
    Who: people with alcohol use disorder
    Effect: folate depletion
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
14. fol-10 · Position of an expert body
    Patients taking methotrexate for cancer should consult their oncologist before taking folate supplements because the supplements could interfere with methotrexate's anticancer effects.
    
    Who: people on methotrexate for cancer
    Effect: possible drug interference
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Folate · checked 2026-09-15
15. fol-11 · Position of an expert body
    Megaloblastic anemia, which is characterized by large, abnormally nucleated erythrocytes, is the primary clinical sign of folate or vitamin B12 deficiency.
    
    Who: people with folate or vitamin B12 deficiency
    Effect: megaloblastic anaemia
    NIH Office of Dietary Supplements, Folate - Health Professional Fact Sheet · checked 2026-09-16 · we read the full text
16. fol-12 · Position of an expert body
    Folate deficiency can also produce soreness in and shallow ulcerations on the tongue and oral mucosa; changes in skin, hair, or fingernail pigmentation; gastrointestinal symptoms; and elevated blood concentrations of homocysteine.
    
    Who: people with folate deficiency
    Effect: oral soreness and ulceration, pigmentation changes, gastrointestinal symptoms, elevated homocysteine
    NIH Office of Dietary Supplements, Folate - Health Professional Fact Sheet · checked 2026-09-16 · we read the full text
17. b12-06 · Position of an expert body
    These neurological symptoms can occur without anemia, so early diagnosis and intervention is important to avoid irreversible damage.
    
    Who: people with B12 deficiency
    Effect: possible irreversible neurological damage
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Vitamin B12 · checked 2026-09-15
18. b12-10 · Observational data
    low vitamin B12...combined with high folic acid...was associated with an almost two to three times higher risk of cognitive impairment.
    
    Who: adults in observational research
    Effect: 2–3× higher risk of cognitive impairment
    NIH Office of Dietary Supplements, Health Professional Fact Sheet: Vitamin B12 · checked 2026-09-15
19. fol-t5 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised controlled trials comparing high-dose with standard-dose folic acid begun before conception · n = 5,144 randomised, 1,488 in the primary miscarriage analysis
    In this population, high-dose folic acid supplementation was associated with a statistically significant reduction in miscarriage risk (RR 0.73; 95% CI: 0.56-0.94; I2=0%). An additional intention-to-treat (ITT) analysis, which included all randomized participants regardless of pregnancy status, showed a nonsignificant reduction in miscarriage risk (RR 0.79; 95% CI: 0.61-1.04). Moreover, the pregnancy rate was significantly higher in the high-dose group compared to the standard-dose group (RR 1.10; 95% CI: 1.02-1.18).
    
    Who: non-pregnant women of reproductive age in three randomised trials; 1,488 of them became pregnant and entered the miscarriage analysis
    Effect: miscarriage risk ratio 0.73 (95% CI 0.56 to 0.94) among those who conceived and clinical pregnancy risk ratio 1.10 (95% CI 1.02 to 1.18); in intention-to-treat analysis of everyone randomised the miscarriage effect fell to 0.79 (95% CI 0.61 to 1.04) and lost significance
    Certainty: low to moderate certainty, only three trials, and the result depends on conditioning on pregnancy: the intention-to-treat estimate that avoids that problem crosses no effect
    Abdelrahman et al., American Journal of Obstetrics & Gynecology MFM, 2026 · checked 2026-09-17 · we read the abstract
20. fol-t6 · Randomised controlled trial(s) · secondary analysis of a placebo-controlled randomised trial of 4.0-5.1 mg folic acid a day from 8-16 weeks' gestation, run at 70 obstetric sites in five countries · n = 462 confirmed twin pregnancies out of 2,464 randomised; 428 analysed
    The rate of preeclampsia was significantly higher in the folic acid group compared to the placebo group in crude analyses (17.2 versus 9.9%; relative risk 1.75 [95% CI 1.06-2.88], p = .029). Multivariable analyses attenuated this effect, rendering it not statistically significant (RR 1.58 [95% CI 0.95-2.63], p = .079).
    
    Who: women pregnant with twins, aged 18 or older, randomised between 2011 and 2015
    Effect: pre-eclampsia 17.2 versus 9.9 per cent, crude relative risk 1.75 (95% CI 1.06 to 2.88, p = 0.029), falling to 1.58 (95% CI 0.95 to 2.63, p = 0.079) and losing significance after adjustment
    Certainty: a subgroup of a larger trial, so not powered for this question; the authors state the suggested elevated risk cannot be confirmed, but nothing here supports high doses for pre-eclampsia
    Wen et al., Journal of Maternal-Fetal & Neonatal Medicine, 2022 · checked 2026-09-17 · we read the abstract
21. fol-t7 · Meta-analysis or systematic review · meta-analysis of individual participant data from randomised controlled trials of prenatal multiple micronutrient supplementation, with iron and folic acid as the control group in most studies · n = 42,618 mother-infant pairs in the multiple micronutrient trials
    In the four-group categorisation, prenatal MMS reduced the risk of preterm-SGA (0·71, 0·62-0·82; p=0·0002) and term-SGA (0·93, 0·89-0·98; p=0·0066).
    
    Who: pregnant women in low- and middle-income countries, mean maternal age 24.3 years, in 14 randomised trials of multiple micronutrients
    Effect: preterm and small for gestational age risk ratio 0.71 (95% CI 0.62 to 0.82, p = 0.0002) and term small for gestational age 0.93 (95% CI 0.89 to 0.98, p = 0.0066) against iron and folic acid
    Certainty: the comparison is multiple micronutrients against iron plus folic acid, not against nothing, so it measures what the other micronutrients add on top of folic acid rather than what folic acid itself does; all trials were in low- and middle-income countries
    Wang et al., Lancet Global Health, 2025 · checked 2026-09-17 · we read the abstract
22. fol-t8 · Meta-analysis or systematic review · systematic review and meta-analysis of randomised trials comparing maternal multiple micronutrient supplementation with iron and folic acid, following infants to 24 months · n = 19 included trials
    Infants born to pregnant women consuming MMS were longer at birth (mean difference: 0.05 cm; 95% CI: 0.02, 0.08 cm) and had higher length-for-age z-score at birth (0.09; 95% CI: 0.06, 0.12), 3 mo (0.09; 95% CI: 0.06, 0.12), and 6 mo (0.04; 95% CI: 0.01, 0.07) of age but not thereafter...MMS reduced risk of stunting (risk ratio [RR]: 0.86; 95% CI: 0.82, 0.91), underweight (RR: 0.86; 95% CI: 0.81, 0.90), small HC (RR: 0.84; 95% CI: 0.79, 0.90), and low MUAC (RR: 0.90; 95% CI: 0.82, 0.99) at 3 mo
    
    Who: infants born to women in 19 randomised trials, measured at birth and at 3, 6, 12, 18 and 24 months
    Effect: length at birth 0.05 cm greater (95% CI 0.02 to 0.08); length-for-age z-score 0.09 higher at birth (95% CI 0.06 to 0.12) and at 3 months, 0.04 at 6 months (95% CI 0.01 to 0.07) and no difference thereafter; stunting at 3 months risk ratio 0.86 (95% CI 0.82 to 0.91)
    Certainty: the differences are small in absolute terms and disappear after six months; the control group received iron and folic acid, so this is the added value of the other micronutrients, not of folate
    Gonçalves et al., American Journal of Clinical Nutrition, 2025 · checked 2026-09-17 · we read the abstract
23. fol-t9 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis of randomised trials of folic acid alongside antifolate antimalarial medication · n = 8 trials with 3,486 participants; 1,062 in the day-7 treatment failure analysis
    People receiving folic acid (alone or with iron) and antifolate antimalarials were less likely to clear malaria parasites (day 3) compared to individuals who did not receive folic acid (RR 0.89; 95% CI 0.84 to 0.95, 4 trials, 929 individuals, moderate-certainty evidence); and had increased risk of treatment failure on day 7 (RR 2.12; 95% CI 1.41 to 3.19, 4 trials, 1062 individuals, moderate-certainty evidence), day 14 (RR 1.97; 95% CI 1.44 to 2.70, 3 trials; 891 individuals; moderate-certainty evidence), and day 28 (RR 1.35; 95% CI 1.21 to 1.51; 4 trials; 1012 individuals, moderate-certainty evidence).
    
    Who: people living in malaria-endemic areas taking antifolate antimalarials, mostly sulfadoxine-pyrimethamine; five of the eight trials were treatment trials, most using folic acid doses above the 1 mg a day upper limit
    Effect: parasite clearance at day 3 risk ratio 0.89 (95% CI 0.84 to 0.95) and treatment failure risk ratio 2.12 at day 7 (95% CI 1.41 to 3.19), 1.97 at day 14 (95% CI 1.44 to 2.70) and 1.35 at day 28 (95% CI 1.21 to 1.51)
    Certainty: moderate-certainty evidence for all four outcomes; most trials used doses above the tolerable upper intake level, so this is not evidence against the standard 400 microgram periconceptional dose, and some trials had unclear or high risk of bias from missing outcome data
    Cochrane Database of Systematic Reviews, 2026 · checked 2026-09-17 · we read the abstract
24. fol-e1-01 · Position of an expert body · rulemaking: FDA states the unit of measure it adopts for folate and the conversion behind it · n = regulatory decision
    As defined by the IOM, mcg DFE is equivalent to mcg food folate + (1.7 x mcg synthetic folic acid) (Ref. 141).
    
    Who: United States food labelling
    Effect: micrograms DFE equal micrograms of food folate plus 1.7 times micrograms of synthetic folic acid, a formula the Institute of Medicine set in 1998; FDA adopted DFE because the plain microgram ignored the difference in bioavailability
    Certainty: a unit of account, not a measurement of anyone's absorption; it also means two foods showing the same number on the label can hold very different amounts of actual folate
    FDA, Food Labeling: Revision of the Nutrition and Supplement Facts Labels (proposed rule), 79 FR 11879 · checked 2026-09-18 · we read the section
25. fol-e1-02 · Position of an expert body · rulemaking: FDA restates the reasoning the Institute of Medicine used · n = regulatory decision
    The IOM developed the new term, DFE, to account for the greater bioavailability of synthetic folic acid that is added to fortified foods or dietary supplements than folate that occurs naturally in foods (food folate).
    
    Who: United States food labelling
    Effect: DFE was created to account for the greater bioavailability of synthetic folic acid added to fortified foods and supplements compared with folate occurring naturally in food
    Certainty: this is the assumption on which the label rests, and the isotope studies in the corpus put the gap much smaller than 1.7 implies; that disagreement is the finding, and neither side of it should be dropped
    FDA, Food Labeling: Revision of the Nutrition and Supplement Facts Labels (proposed rule), 79 FR 11879 · checked 2026-09-18 · we read the section
26. fol-e1-03 · Randomised controlled trial(s) · four-week randomised dietary intervention with four arms, carbon-13-labelled folic acid given to everyone as an internal tracer · n = 72 adults
    Bioavailability of food folate relative to that of folic acid was 78% (95% CI: 48%, 108%) according to [(13)C(11)]-labeled folate and 85% (52%, 118%) according to changes in serum folate concentrations.
    
    Who: 72 healthy adults; one group ate a high-folate diet of 369 ug food folate a day with placebo, three groups ate 73 ug food folate plus 92, 191 or 289 ug folic acid
    Effect: relative bioavailability of food folate 78% (95% CI 48% to 108%) by labelled folate and 85% (52% to 118%) by change in serum folate
    Certainty: the confidence intervals are very wide and both include 100%, so this does not prove food folate equals folic acid; it does mean the 1.7 factor is not securely below the data. This is aggregate bioavailability from a mixed diet, which is a different question from any single food. Full text not available to us (not in PMC)
    Winkels et al., American Journal of Clinical Nutrition, 2007 · checked 2026-09-18 · we read the abstract
27. fol-e1-04 · Randomised controlled trial(s) · four-week randomised dietary intervention, authors' conclusion · n = 72 adults
    The consumption of a diet rich in food folate can improve the folate status of a population more efficiently than is generally assumed.
    
    Who: 72 healthy adults
    Effect: the aggregate bioavailability of folates from fruit, vegetables and liver is approximately 80% of that of folic acid
    Certainty: one study with wide intervals, and it does not touch the separate question of whether folic acid prevents neural tube defects, which is settled and is about a supplement taken before conception. Full text not available to us (not in PMC)
    Winkels et al., American Journal of Clinical Nutrition, 2007 · checked 2026-09-18 · we read the abstract
28. fol-e1-05 · Randomised controlled trial(s) · single-centre randomised four-treatment four-period crossover study in Latin square order, stable isotope dilution assays, 14-day washout between foods and folic acid saturation beforehand · n = 24 subjects, 4 test items each
    When comparing the ratio estimates of AUC and C(max) for the different test foods, highest bioavailability was found for spinach followed by that for wheat germs and Camembert cheese.
    
    Who: 24 healthy adults, each receiving all four test items
    Effect: ranking of relative bioavailability by AUC and Cmax of 5-methyltetrahydrofolate over 12 hours: spinach highest, then wheat germ, then Camembert; median time to peak was shortest for spinach; variability was moderate for spinach and very high for wheat germ
    Certainty: a 12-hour biokinetic comparison in 24 people, not an outcome study; the very high variability for wheat germ means its position in the ranking is unstable. Full text not available to us (not in PMC)
    Food & Function, 2015 · checked 2026-09-18 · we read the abstract
29. fol-e1-06 · Randomised controlled trial(s) · randomised crossover stable-isotope study, authors' conclusion · n = 24 subjects
    The results underline the dependence of folate bioavailability on the type of food ingested.
    
    Who: 24 healthy adults
    Effect: the results underline that folate bioavailability depends on the type of food ingested, which the authors set against the general assumption of a fixed 50% figure
    Certainty: three foods in 24 people cannot replace a conversion factor, it can only show that one number will not fit all of them. Full text not available to us (not in PMC)
    Food & Function, 2015 · checked 2026-09-18 · we read the abstract
30. fol-e1-07 · Meta-analysis or systematic review · systematic review with random-effects meta-analysis and GRADE rating of the substitution question · n = 65 publications covering 61 unique primary studies and 71 analytic contributions; the pooled substitution question rests on 3 trials
    5-MTHF substitution reduced plasma UMFA (SMD +0.99; 95% CI +0.20 to +1.79; I2 = 30%; k = 3; GRADE low).
    
    Who: participants in trials comparing equimolar 6S-5-methyltetrahydrofolate against folic acid
    Effect: 5-MTHF substitution reduced plasma unmetabolised folic acid, SMD +0.99 (95% CI +0.20 to +1.79), I-squared 30%, GRADE low
    Certainty: three trials, GRADE low, and reducing a blood marker is not the same as producing a health benefit; the two trials that measured red cell folate pointed in opposite directions and could not be pooled
    Nutrients, 2026 · checked 2026-09-18 · we read the abstract
31. fol-e1-08 · Meta-analysis or systematic review · systematic review with secondary pooled analyses by MTHFR 677 genotype · n = secondary pool within a review of 61 primary studies
    This is a research question rather than a basis for genotype-guided supplementation in practice, and MTHFR 677TT carriers are a plausible early target population for confirmatory substitution trials.
    
    Who: trial participants stratified by MTHFR 677 genotype; TT homozygotes are about 10% of European-ancestry populations and more in Mexican and Mediterranean-ancestry ones
    Effect: TT participants may show a larger homocysteine response to folic acid, but the pattern was not consistent within every trial and no pooled genotype trend was estimated
    Certainty: the reviewers state explicitly that this is a basis for confirmatory trials in TT carriers, not for genotype-guided supplementation in practice; it is quoted because MTHFR testing is sold as if the opposite were true
    Nutrients, 2026 · checked 2026-09-18 · we read the fulltext
32. fol-e1-09 · Meta-analysis or systematic review · systematic review, discussion of the identifiability problem across the observational literature · n = 24 observational contributions within the review
    A major methodological limitation of this field is that effects of UMFA cannot readily be identified independently of total circulating folate.
    
    Who: the observational studies relating unmetabolised folic acid to health outcomes
    Effect: folic acid intake raises unmetabolised folic acid and total folate together and monotonically, so in observational data the two are strongly collinear; none of the directly-measuring studies identifies an effect specific to unmetabolised folic acid
    Certainty: this is the reviewers' methodological verdict, and it cuts both ways: it is as much an argument against alarm about unmetabolised folic acid as against reassurance
    Nutrients, 2026 · checked 2026-09-18 · we read the fulltext
33. fol-e1-10 · Meta-analysis or systematic review · systematic review, discussion of measurement properties · n = within a review of 61 primary studies
    Unmetabolized folic acid also constitutes no more than 1–2% of serum total folate [76], making measurement error proportionally large relative to its dynamic range, and a single measurement conflates dose with time since last ingestion; in the Boston Birth Cohort, time since last FA exposure was not recorded [33,76].
    
    Who: studies measuring circulating unmetabolised folic acid
    Effect: unmetabolised folic acid is no more than 1-2% of serum total folate, so measurement error is proportionally large against its dynamic range, and one measurement conflates dose with time since last ingestion; in the Boston Birth Cohort that timing was not recorded
    Certainty: a statement about measurement, not about safety; it explains why the observational signals on this marker are inconsistent and unreplicated
    Nutrients, 2026 · checked 2026-09-18 · we read the fulltext
34. fol-e1-11 · Meta-analysis or systematic review · systematic review, narrative synthesis of the observational stream classified by exposure metric · n = 24 observational contributions
    Observational UMFA-outcome associations were inconsistent, confounded by total folate, and unreplicated.
    
    Who: observational studies of unmetabolised folic acid, total folate, folic acid intake and population fortification
    Effect: observational associations between unmetabolised folic acid and outcomes were inconsistent, confounded by total folate, and unreplicated
    Certainty: an absence of a reliable signal, not a demonstration of safety; the review's own conclusion is that the evidence cannot support an equivalence or non-inferiority claim in either direction
    Nutrients, 2026 · checked 2026-09-18 · we read the abstract
35. fol-e1-12 · Randomised controlled trial(s) · 24-week randomised controlled trial of a prenatal multivitamin containing 6S-5-methyltetrahydrofolate against one containing folic acid, started in the second trimester · n = 80 randomised, 62 completers
    By week 24 of the study, 5-MTHF and total folate concentrations in maternal and cord serum and the placenta did not differ between groups.
    
    Who: 80 first-trimester pregnant participants randomised, 62 completed, 31 per group
    Effect: 1,000 ug DFE of 6S-5-MTHF against 1,330 ug DFE of folic acid: no difference in 5-MTHF or total folate in maternal serum, cord serum or placenta at week 24; detectable unmetabolised folic acid in 7% against 31% of mothers (P < 0.01), and lower in the placenta
    Certainty: 62 completers and no clinical outcomes measured: the trial compares blood chemistry, not babies, and the authors say more research is needed on how folate form affects fetal health
    Frontiers in Nutrition, 2026 · checked 2026-09-18 · we read the abstract
36. fol-e1-13 · Randomised controlled trial(s) · 24-week randomised controlled trial with maternal blood at weeks 0, 12 and 24 and cord blood and placenta at delivery · n = 62 completers, 31 per group
    There were significantly fewer participants (7% vs. 31%) with detectable UMFA and lower average UMFA concentrations in maternal blood of the MTHF-MVI group vs. the FA-MVI group (p < 0.01).
    
    Who: pregnant participants from the second trimester onward
    Effect: detectable unmetabolised folic acid in 7% of the methylfolate group against 31% of the folic acid group with lower average concentrations (P < 0.01); only 2 cord samples had detectable unmetabolised folic acid, both from the folic acid arm
    Certainty: the doses were not equal, 1,000 against 1,330 ug DFE, so part of the difference is dose rather than form; and whether detectable unmetabolised folic acid does any harm is exactly the question the 2026 systematic review says nobody can currently answer
    Frontiers in Nutrition, 2026 · checked 2026-09-18 · we read the abstract