Putting It Back

Folate, Folic Acid, Folinic Acid, Methylfolate: Four Names, Four Molecules

The confusion starts with the word itself. Vitamin B9 is not one molecule but a family of them, and the labels use the names interchangeably, as though they were the same. They are not.

Folate is the natural family. It is found in leafy greens, legumes, liver and eggs, and in food it arrives in several reduced, chemically delicate forms. Whatever form we eat, our bodies convert it, ultimately, into 5-methyltetrahydrofolate (5-MTHF), the form that circulates in the blood.

Folic acid is the odd one out. As the researchers Steven Bailey and June Ayling put it in a widely cited 2009 paper, it is a synthetic, oxidised form that is not significantly found in fresh natural foods. It is stable, cheap and keeps for a long time, which is exactly why it is used in supplements and in fortified flour. But it is a form the body cannot use as it is. It has to be reduced, step by step, into active forms, and the first step belongs to an enzyme called DHFR, of which more shortly. By weight, it is better absorbed than food folate: 1 microgram of folic acid counts as 1.7 micrograms of dietary folate equivalents.

Folinic acid (5-formyltetrahydrofolate, known as leucovorin in medicine) is already reduced, so it does not need DHFR. In a 2022 report of two patients with cerebral folate deficiency, the authors noted that folinic acid is efficiently metabolised to 5-MTHF in the intestine and liver. It is used mainly as a medicine in specific conditions.

L-methylfolate (5-MTHF) is the active form that dominates the bloodstream and the one that enters the brain. It is sold as a supplement, usually as a stabilised salt.

The last row deserves more weight than it usually gets. 5-MTHF is the form the blood already carries, and it needs no DHFR to be used. In a randomised trial in women of childbearing age, 416 micrograms a day raised red blood cell folate more than 400 micrograms of folic acid did, although a lower-dose trial found the two similar. In a single-dose study, 5-MTHF produced higher levels of 5-MTHF in the plasma, and only folic acid produced a transient peak of unmetabolised folic acid. What the active form lacks is not evidence that it works on folate status. It is a trial on birth defects, and until that exists, the strongest evidence for preventing them remains folic acid’s.

How the Flour Lost Its Vitamins

For most of history, bread was made from whole grain, ground between stones, and eaten soon after. The trouble with whole-grain flour is that it does not keep. The germ of the wheat, the living embryo of the seed, holds oils, which go rancid. When steel roller mills spread through North America and Europe in the second half of the nineteenth century, they solved that problem by separating the grain into its parts and discarding the bran and the germ, leaving only the starchy endosperm. White flour was whiter, finer and, because the oils were gone, nearly indestructible. A miller could now store it, ship it and sell it far from where it was grown.

The price was paid in nutrition. Removing the bran and germ strips out a good share of the fibre, iron, B vitamins and essential fats. In populations that came to depend on white flour and refined grains, the deficiency diseases that followed were pellagra, caused by a lack of niacin, and beriberi, caused by a lack of thiamin, along with anaemia. The response, in the United States, came in 1941, when health officials alarmed by the state of young men arriving at military recruitment centres moved to enrich white flour: to be sold as enriched, it was to have thiamin, riboflavin, niacin and iron added back. Pellagra and beriberi all but disappeared. It was an extraordinary public-health success, and also a strange one: the solution to a problem created by taking nutrients out was to put a few of them back, in synthetic form, rather than to restore the grain.

A thought follows naturally from this, and I have heard it stated as fact: that milling must also have given rise to spina bifida. It is a reasonable hypothesis. Milling removes a substantial share of the folate in whole grain, according to the milling industry’s own summaries, and low folate status around conception is a major risk factor for neural tube defects, so populations that came to rely on refined flour may well have had lower folate status and a higher risk. But I could not find historical data to confirm it. Birth registries did not exist until the 1960s, so nobody counted neural tube defects before and after the roller mills arrived, and other foods, such as leafy greens, legumes and liver, also supply folate and varied widely between regions and social classes. It remains a plausible hypothesis, not a documented fact. What is documented is how long the wait was. The B vitamins were put back in 1941. Folic acid was not added to the US standard until 1998, 57 years later, because the first randomised proof that it prevents neural tube defects, in women at high risk, only arrived in 1991.

And there is a second honest complication for those of us who would prefer to return to whole grain. A whole-grain loaf is a better food in many ways, but whether diet alone could deliver the protection that was shown is less certain than it sounds. A review in the Archives of Disease in Childhood made the case that the intake used in the trial, 4 milligrams a day, cannot be reached from dietary change alone, so supplements or fortified food are necessary. That is the argument that has persuaded governments. But a trial dose that is far above what diets normally deliver is not the same as saying that a good diet does nothing, and the section after next looks at what the older numbers suggest.

Why Folic Acid, and Who Has Fortified

The case for folic acid rests on a handful of trials, and above all on one from 1991. The Medical Research Council Vitamin Study randomised 1,817 women who had already had a pregnancy affected by a neural tube defect, a group at roughly ten times the usual risk, to folic acid, other vitamins, both or neither. Of the 1,195 pregnancies with a known outcome, 6 of the 593 among women taking 4 milligrams of folic acid a day before conception (1.0 percent) were affected by a neural tube defect, which includes spina bifida and anencephaly, against 21 of the 602 in the other groups (3.5 percent). That is a relative reduction of about 72 percent. In absolute terms, it is a fall of about 2.5 percentage points, or roughly 25 fewer affected pregnancies per 1,000, which by my arithmetic means that about 40 women had to take folic acid for one recurrence to be avoided. The neural tube closes about 28 days after conception, often before a woman knows she is pregnant, and nearly half of all UK pregnancies are unplanned, so a supplement that has to be started in advance reaches too few of the women who need it. Governments reached for the one intervention that reaches everyone: flour.

That result is strong, but it helps to know what it does and does not show. The trial was well designed: because the women were split into four groups, the researchers could tell the effect of folic acid apart from the effect of the other vitamins. But there were very few cases to count, 27 recurrences in all, and the trial was stopped early because the first results looked so clear, which can make an effect seem bigger than it is. All the women had already had an affected pregnancy, so they were at much higher risk than most. And they took 4 milligrams a day, about ten times the 400 micrograms now recommended and far more than flour provides. No lower doses were tested.

For women who had never had an affected pregnancy, the evidence is thinner. It rests mainly on one trial, from Hungary in 1992. The women in the test group took a multivitamin containing 0.8 milligrams of folic acid, and the women in the comparison group took a trace-element supplement instead. Because the multivitamin contained other vitamins as well, the trial cannot show that folic acid alone made the difference. It recorded only six neural tube defects, all of them in the comparison group, which is roughly 1 in 400 of those women, and none in the multivitamin group.

When five randomised trials were pooled, a Cochrane review found that folic acid supplements reduced neural tube defects, with a risk ratio of 0.31, a relative reduction of about 69 percent. It noted that the trials were done before flour fortification was widespread, and in women of Caucasian and Indian origin, so the results may not apply equally everywhere. For first-time occurrences, the effect was not statistically significant, which means there were too few cases to rule out chance.

One distinction matters throughout this piece. A relative reduction is the proportional fall in risk. An absolute reduction is the difference in how many people are actually affected. In the first trial, the 72 percent relative reduction meant a fall from 3.5 percent to 1.0 percent among women at high risk. If the same proportional effect applied to women at the general population’s risk, which is an assumption, a risk of about 1 in 1,000 would fall by roughly 0.7 in 1,000.

The effect of fortification, meanwhile, comes from before-and-after comparisons of whole populations, not from trials. In relative terms, Canada reported a 46 percent fall, from 1.58 to 0.86 affected births per 1,000, which is 0.72 fewer per 1,000, or about one fewer for every 1,400 births. The US figure was a 28 percent reduction in the CDC’s analysis, and the UK government expects a 15 to 22 percent reduction. If the UK’s starting rate is close to Europe’s roughly 1 per 1,000 pregnancies, that works out at about 0.15 to 0.22 fewer per 1,000, or one fewer affected pregnancy for every 4,500 to 6,700 pregnancies. Those are meaningful reductions in a serious condition. They are smaller than the headline figure of 72 percent, which is itself a relative reduction in a high-risk group, and they are measured in a different way.

The United States required folic acid in enriched grain products from 1998. An early analysis in JAMA found a 19 percent relative fall in neural tube defect births, from 37.8 to 30.5 per 100,000 live births, and a later CDC study of 19 surveillance programmes found a 28 percent relative reduction in anencephaly and spina bifida, about 1,300 cases prevented a year. Canada moved in the same year, and recent counts put the number of countries that fortify cereal grains with folic acid at around 80. Most of Europe has not. The UK, which held back for more than two decades over fears about older people with undiagnosed vitamin B12 deficiency, is now joining the list, and its government expects to prevent around 200 cases of neural tube defects a year.

Those are the facts on one side of the ledger, and they are substantial. A population measure, though, is one that reaches everyone, including the many people who do not need it and a few who might be harmed by it. The rest of this piece is about those people, and about what the science says on the other side.

Before the Fortification: What the Older Numbers Say

How common were neural tube defects before white flour, before fortification, before anyone was counting? The honest answer is that we do not know. There are no reliable population figures from before the twentieth century, and the first careful registries date from the 1960s. One of the earliest, a study of births in Belfast between 1964 and 1968, reported rates above 4 per 1,000 births. Those rates were not a constant of nature. Across Britain and Ireland they fell steeply from the 1970s. EUROCAT, the European registry network, describes a well-documented steep fall in prevalence in the UK and Ireland before the 1990s, partly due to prenatal diagnosis and termination of affected pregnancies, and one summary of English and Welsh registry data reports total prevalence, terminations included, falling from about 3.7 per 1,000 births in 1964 to 1.6 in 1990, a relative fall of more than half.

Termination explains part of the fall in live births. It cannot explain a fall in total prevalence, which counts terminated pregnancies too, so some of the improvement must have come from elsewhere. Researchers including Elwood and Scott suggested that better nutrition, vitamins included, played a part, and neural tube defects are known to carry a socioeconomic gradient. So it is plausible, and to my mind likely, that better overall nutrition and living conditions explain a share of the historical improvement. The data cannot prove it.

What about places with no fortification today? In Europe, registries that count terminations too put the total prevalence at around 1 per 1,000 pregnancies, about 5,000 pregnancies a year. EUROCAT describes it as relatively stable despite years of advice to take supplements. Only a minority of European women, by one estimate 10 to 30 percent, take the recommended dose before conception, and 70 to 80 percent of affected pregnancies end in termination. After fortification, Canada reported a fall from 1.58 to 0.86 per 1,000 births, a relative reduction of 46 percent, and the United States from 0.76 to 0.56 per 1,000, about 26 percent, or 0.2 fewer per 1,000 births. Those figures count births and not pregnancies, so they are not strictly comparable with Europe’s, but they point in the same direction.

My reading is that the story has two chapters. In the first, across the twentieth century, rising living standards and better nutrition seem to have lowered the baseline, perhaps substantially. In the second, a further fall followed fortification where it was introduced, while in Europe, where the main prevention has been advice, no comparable fall has appeared. That does not make food irrelevant. It suggests that diet moves the risk, and that the last stretch of prevention needs more folate than most diets supply. That is a hypothesis that fits the data, not something the data have proven.

The Slow Enzyme: DHFR and What Is Left Over

To be used, folic acid must be reduced to tetrahydrofolate, and the enzyme that does this is dihydrofolate reductase, or DHFR. Here is where the biochemistry gets interesting, and where much of the modern concern comes from.

Much of what we know about folic acid was learned in rats, whose DHFR is fast. In 2009, Bailey and Ayling at the University of South Alabama measured the activity of the enzyme in fresh human liver, taken from organ donors or from surgery, and compared it with rat liver. In the six human livers they studied, DHFR activity was on average less than 2 percent of that in the rat, about 56 times lower, and it varied almost five-fold from one human to the next. Their conclusion was careful: the extremely low rate of conversion suggests that the benefit of high doses will be limited by saturation of the enzyme, especially in people with lower than average activity, and the result is consistent with reports of unmetabolised folic acid in plasma and urine. The point worth underlining, because it is often presented as a problem of genetic variants, is that this is a feature of human biology in general. It is not limited to people with MTHFR variants. It is slow in all of us, and more so in some.

It is slow in all of us, and more so in some.

The gut shows the same limitation. In a small study from the University of East Anglia and the Quadram Institute, in patients with a liver shunt that gave access to blood in the portal vein, 15 minutes after a dose of labelled folic acid, 80 percent of the labelled folate in the portal blood was still unmodified folic acid. The authors concluded that humans rely on the liver to reduce folic acid, even though liver DHFR is low and highly variable, and that chronic exposure may saturate it.

What happens when the enzyme is saturated is that folic acid itself appears in the bloodstream, unconverted. This is known as unmetabolised folic acid, or UMFA. It becomes detectable in the blood after single doses of more than about 200 micrograms. Bailey described it this way: as the dose rises, one reaches a point at which DHFR cannot go any faster, and unconverted folic acid begins to accumulate in the blood.

So what does it do there? Honestly, we do not know. The UK’s own advisory committee, SACN, concluded in its 2006 risk assessment that there were insufficient data on the implications of unmetabolised folic acid in circulation, and, as far as I can tell, the question has not been settled since. Some studies have reported an association between higher levels and lower natural killer cell activity in older women, but Bailey himself said it was too soon to say what long-term effect higher doses of unmetabolised folic acid might have. The meaning of its presence in plasma is, in the words of one pharmacokinetic study, controversially discussed. A candid summary is that UMFA is real, is measurable, and has consequences that remain unproven in either direction.

MTHFR: Who Carries It, and What It Really Changes

If DHFR is the slow enzyme in everyone, MTHFR is the famous one. The gene produces an enzyme, methylenetetrahydrofolate reductase, that performs a different step further down the road: it converts one reduced folate into 5-MTHF, the form that circulates in the blood. The best-known variant, C677T, makes the enzyme less active and more fragile. People with one copy of the variant (heterozygous) have roughly 65 percent of normal activity, and those with two copies (homozygous, TT) around 30 percent.

How many people does this concern? Around 8 to 12 percent of people of European ancestry are homozygous and about 40 percent heterozygous, which is how the figure of roughly half of the population arises. Worldwide, a meta-analysis estimated the T allele at 24 percent and the TT genotype at 7.7 percent, with the frequency lowest in people of African ancestry and highest in Europeans. So it is true that a large share of us carry a variant. It is not true, though, that these people cannot metabolise folic acid. MTHFR is not the enzyme that processes folic acid in the first place. That task falls to DHFR, which is slow in everyone. What the MTHFR variant reduces is the later conversion into 5-MTHF, which in the homozygous case tends to go along with lower blood folate and mildly raised homocysteine.

And here is the part that complicates the popular story. In people with the TT genotype, the raised homocysteine can be corrected by folic acid, and maternal folic acid supplementation appears to prevent neural tube defects in part by compensating for the reduced MTHFR activity. In other words, the people with the variant are not obviously the ones with the most to fear from folic acid, and the fall in neural tube defects after fortification happened in whole populations, in which carriers are common. A reasonable summary is that having an MTHFR variant does not make folic acid dangerous. It does mean that your levels of the active form may be lower for a given intake, and that matters for how much you get from food.

The counter-argument comes mostly from the functional-medicine world, and its best-known voice is Dr Ben Lynch, a naturopath and author of Dirty Genes. His core claims are consistent across his interviews and podcasts. Folic acid blocks folate receptors; it binds preferentially to receptors designed for methylfolate; it is unnatural to the body; and people consume too much of it, which is why he advises minimising folic acid in favour of methylfolate. The first claim is based on genuine biochemistry, which I turn to next. The conclusion that people with MTHFR variants should avoid folic acid is the part that the clinical evidence does not currently support, and I will say so plainly.

The Brain and the Doorway

The most striking claim in the circles around MTHFR is about the brain: that folic acid occupies the doorway through which active folate has to enter, so that methylfolate is left outside. It deserves a careful hearing, because it is partly true.

Folate does not simply diffuse into the brain. It crosses mainly at the choroid plexus, the tissue that makes cerebrospinal fluid (CSF), through a protein called the folate receptor alpha. Its job is to capture 5-MTHF in the blood with high affinity and carry it across in an energy-dependent process that leaves the CSF with more 5-MTHF than the plasma has. Mutations or antibodies that disable this receptor cause a rare condition, cerebral folate deficiency, in which CSF folate is low even when blood folate is normal.

Now the biochemistry. The receptor has a higher affinity for folic acid than for 5-MTHF. In principle, therefore, folic acid can act as a competitive inhibitor, and clinicians have long discouraged folic acid in cerebral folate deficiency, in favour of folinic acid. A 2022 case report described two patients with that condition who were given high-dose folic acid: their CSF 5-MTHF did not normalise, and their blood showed a gap between total folate and 5-MTHF that the authors attributed to unmetabolised folic acid. They concluded that excess folic acid impaired transport across the barrier. The authors also noted that folic acid cannot be efficiently converted in the brain, where DHFR activity is extremely low. A simplification in the popular version is that folic acid merely sits at the door. The receptor binds folic acid avidly, and what reaches the brain is very hard to use.

Here is what we do not know. Those were two patients with a rare disorder on high doses, and the report itself begins by describing the inhibition as, until then, theoretical. I found no study showing that the amounts of folic acid in fortified flour reduce folate delivery to the brain in healthy people. Nor does the evidence show that fortification worsens attention or mood. Dr Lynch himself, speaking about ADHD, says there “may be” a link with the synthetic folate in food. That is a hypothesis worth testing, and no more than that. Most people eat plenty of other folates, and the blood carries a great deal of 5-MTHF that the receptor can bind. What we have is a plausible mechanism with some evidence at high doses in people who are already vulnerable, and no evidence yet at the doses most of us meet in bread.

The Other Worries, Weighed

The doorway and the enzyme are newer concerns. The older ones have been argued over for decades, and they are the ones that kept British flour unfortified.

Hiding vitamin B12 deficiency. B12 deficiency causes a particular kind of anaemia and, if untreated, irreversible damage to the nerves. Folic acid can correct the anaemia, and so the blood count looks better while the nerve damage proceeds. This is the worry that led the Food Standards Agency to reject earlier advice to fortify, because B12 deficiency is common in older adults. The UK government’s own impact assessment records that SACN, its scientific advisory committee, reviewed the evidence on masking and on cognitive decline in older people, and found no effect of folic acid supplementation on cognitive decline, and SACN’s advice remains in favour of mandatory fortification alongside controls on voluntary fortification and clear guidance on supplements. Supporters point to American data showing no increase in B12 deficiency after fortification began. Critics reply that undiagnosed B12 deficiency is common in older age and that nobody has been monitoring closely. SACN itself had asked, in earlier years, for better ways of diagnosing it.

Cancer. This is the most contested. A large trial of 1 milligram a day of folic acid in people with a history of colorectal adenomas suggested a possible increase in the risk of advanced lesions and in the number of adenomas. Observers have also pointed to a reversal of the downward trend in colorectal cancer in the US and Canada around the time of fortification, though that trend is open to other explanations. SACN reviewed prostate, breast, colorectal and overall cancer risk, along with the effects of unmetabolised folic acid, and still recommended fortification. The trial dose of 1 milligram a day was several times the roughly 100 to 130 micrograms a day by which American intakes rose after fortification. It is also true that nobody has run a trial at the dose of flour fortification across a lifetime.

The immune system. The observation about natural killer cells, mentioned above, came from a study of older women with higher levels of unmetabolised folic acid. It is an association in a small group, not evidence of harm.

I come away with the feeling that the honest position is uncomfortable for both camps. The benefit, fewer babies born with devastating birth defects, is real and measured. The harms on the other side are mostly hypotheses, a few of them biologically plausible and some supported by high-dose data, and none established at the dose in flour. A policy made on the strength of a measured benefit and a set of unmeasured risks is not necessarily wrong. But it does mean that the monitoring that was promised matters more than usual.

What Changes in the UK on 13 December

The British rule is now close. Under the Bread and Flour (Amendment) Regulations 2024, all non-wholemeal wheat flour milled in the UK must, from 13 December 2026, contain 250 micrograms of folic acid per 100 grams, in addition to the calcium, iron, niacin and thiamin already required. Wholemeal flour is exempt, the rule covers flour from common wheat only, and mills producing under 500 tonnes of white flour a year may opt out, to spare small producers the burden of compliance. The government expects the change to prevent around 200 cases of neural tube defects a year, a relative fall of 15 to 22 percent in affected pregnancies. Some millers are moving early: one independent miller began fortifying its white flours in June 2026. Because the flour goes into bread, biscuits, pastry and pasta, folic acid is likely to turn up across the diet, and Ireland’s food safety authority has already warned that products made with UK white flour could reach the Irish market.

Who does this affect, beyond pregnant women? Everyone who eats white flour, and for a few groups, health advice is more cautious. NHS guidance on folic acid as a medicine tells people to speak to a doctor first if they have vitamin B12 deficiency or pernicious anaemia, or cancer, and the version I saw also lists haemodialysis and a heart stent. Two points are needed to read that list properly. The first is that it concerns tablets taken in medicinal doses, usually 400 micrograms or more a day, and not the smaller amount that arrives in bread. The second is that the caution does not apply uniformly. Folic acid tablets are licensed in the UK to treat anaemia in patients on kidney dialysis, so for renal patients the issue is more likely to be dose and supervision than a blanket ban. The UK Clinical Pharmacy Association notes that the caution about coronary stents appears to date from a past trial in which folate was used to prevent re-narrowing of the stent, which did not work and may have made things worse, and that use for treating folate deficiency should not be a concern.

The government’s impact assessment lists the effect on people who are unable or choose not to consume folic acid, and on at-risk groups such as those with B12 deficiency, among the questions it considered. I could not find a published study of what fortification-level exposure does in kidney patients, people with cancer or those with undiagnosed B12 deficiency, and I think that gap is itself worth stating. Anyone in these groups should ask their own doctor or renal dietitian, not rely on a general article. And a practical point for everyone: the added nutrients are listed on the label after the word flour, as folic acid, so it can be spotted, and because wholemeal flour is exempt, choosing it is one way to avoid it.

California and the Tortilla

The American story has its own loose end. When the FDA mandated folic acid in 1998, it covered enriched wheat flour, rice and corn flour, but not corn masa, the nixtamalised corn dough that goes into tortillas, tamales and pupusas. The omission mattered, because Hispanic communities eat a great deal of masa and face a higher risk of neural tube defects. A CDC analysis in 2009 noted that Mexican-Americans carry a risk of severe brain and spinal birth defects that is 30 to 40 percent higher in relative terms and estimated that fortifying masa could raise their folic acid intake by nearly 20 percent. In April 2016, the FDA finally allowed manufacturers to add folic acid to masa, but did not require it. Two years later, a study in JAMA found that only about 10 percent of the masa flour tested contained folic acid, and none of the corn tortillas.

Which brings us to California. The mandate is not coming, it is already here. Assembly Bill 1830, introduced by the physician and Assemblymember Joaquin Arambula and signed in September 2024, took effect on 1 January 2026. Corn masa flour manufactured, sold or used in foods in California must contain 0.7 milligrams of folic acid per pound, which works out at about 150 micrograms per 100 grams, a little under the UK’s rate, and wet masa products 0.4 milligrams per pound. The addition must be declared on the label. Cottage food operations and restaurants that make masa on site are exempt, and Alabama has since passed a similar law. The CDC has estimated that fortifying masa at the same level as other enriched grains would prevent about 40 additional cases of neural tube defects a year in the United States.

The market logic is straightforward. A company that sells across the country is unlikely to want a separate production line for one state, so a Californian requirement can quietly become a national one. I could not find data on how many manufacturers are actually reformulating for the whole country, so I would call that a plausible expectation, not a documented fact. One large producer, Gruma, the parent of Mission Foods, says it began adding folic acid to its masa products in 2016 and supports the new laws. The consequence for anyone who prefers to avoid synthetic folic acid is that the corn tortilla, which until recently was usually unfortified, is becoming harder to find without it.

Is It Right to Put It in Everyone’s Bread?

Everything so far has been about biology. There is also an ethical question, and it is a serious one: what does it mean to add a synthetic compound to a staple food, so that every person, from a toddler to a great-grandmother, takes it in every day, for life, whether or not they need it, and without being asked?

The framework most often used in Britain to think about this comes from the Nuffield Council on Bioethics, whose public health guidance asks of any measure: is it proportionate, is the risk scientifically assessed, is it fair and does it respect consent, is it transparent, and are the benefits worth the costs. It also describes an intervention ladder. At the bottom rung, a government informs people. At the top, it removes the choice. Mandatory fortification of a staple is close to the top, which means the justification has to be strongest and the evidence most complete.

The case for is real. The harm is grave and falls on someone who cannot consent to anything, an unborn child. Nearly half of UK pregnancies are unplanned, only a minority of European women take a supplement before conception, and the risk of neural tube defects is not spread evenly: it is higher in poorer communities and, in the US, among Hispanic women. Fortification reaches the people that advice does not, and by the CDC’s estimate it prevents about 1,300 cases a year in the US.

The case against is also real. Nobody can opt out of flour in a world where bread, pastry, biscuits and pasta all contain it, and individuals and their doctors lose the ability to tailor their intake to their genes, diet or health. The benefit goes to a small group, mostly unborn babies and women of childbearing age, while the exposure goes to everyone, including men, children and older adults who gain little from it. One recent review sums up the objections as the absence of consent, the inability to opt out, the uneven distribution of risk and the lack of transparency about possible harm, and notes that no country has reversed mandatory fortification once it was introduced. Ireland’s food safety authority put its own plan on hold in 2009 until more evidence on safety was available.

Children and older adults deserve a closer look. Folic acid has an upper intake level, 1,000 micrograms a day for adults and only 300 for children aged one to three. In the US, fewer than 3 percent of adults exceed theirs. Among children, the ones who go over are mostly those taking folic acid supplements: in a national survey, almost every child aged 1 to 8 who got more than 200 micrograms from supplements was above the limit for their age, and a Canadian analysis found that one to four percent of children were over the limit when foods and supplements were combined, and suggested removing folic acid from supplements designed for young children. In other words, for children the problem seems to come mostly from additional sources, not from flour alone. Australian risk modelling at a comparable fortification level found that none of the people aged 70 and over would exceed the limit. For older adults, the specific worry is vitamin B12, which I discussed above.

And then there is for life. The United States and Canada have now lived with mandatory fortification for more than 25 years, which is a very large natural experiment. But as far as I can tell it was monitored mainly for birth defects and for blood folate, and was not designed to detect small effects on cancer or on the brain in old age, and the trials that raised these questions lasted a few years, not a lifetime. Absence of evidence of harm is reassuring. It is not the same as evidence of safety.

Absence of evidence of harm is reassuring. It is not the same as evidence of safety.

My own conclusion is that the case for fortification is strong enough to be taken seriously, and the ethical cost is high enough that it should come with conditions. Choice should be preserved where it can be, with unfortified flour kept on the shelf and clear labelling, as the UK’s exemption for wholemeal and for the smallest mills partly does. The monitoring that was promised, of B12 status, of unmetabolised folic acid and of long-term outcomes in older people, should actually happen, with the results published. The policy should carry a review date, so that it can be changed as the evidence comes in. And the question of whether the active form could do the same work with fewer unknowns should be put to the test, not left as a footnote.

What You Can Do With This

None of this is a reason to alarm anyone, and some of it is a reason to be careful in the other direction. A few principles come out of the evidence.

If you might become pregnant, do not drop folic acid because of this article. The evidence for preventing neural tube defects is for folic acid, and the NHS advises all women who could become pregnant to take 400 micrograms a day, starting before conception. So if your prenatal contains folic acid, keep taking it. But it is also reasonable to look into a prenatal that contains the active form, 5-MTHF, on its own or alongside folic acid. In the trials I found, 5-MTHF raised blood folate as well as folic acid or better, it does not rely on the slow DHFR step, and it did not leave unmetabolised folic acid in the blood. The honest caveat is that those trials measured folate levels and not birth defects, and a 2024 review concluded that clinical studies of 5-MTHF for preventing neural tube defects are lacking. That is a good question to put to a doctor or midwife, particularly if you carry an MTHFR variant, and the sensible way to make a change is with their guidance, not instead of it.

Eat folate in food anyway. Leafy greens, legumes, eggs and liver deliver the natural folates, together with the other nutrients that travel with them. The history above suggests that overall nutrition matters. We do not know whether food alone can close the remaining gap in pregnancy, but it is the best baseline for everyone.

Know your own situation. If you have, or might have, vitamin B12 deficiency, the way to find out is a blood test, not folate guesswork. People with kidney disease, cancer, or on medicines that act on folate should ask their own doctor or specialist team. And remember that ordinary serum folate tests do not tell the forms apart. I found no guidance that says testing for unmetabolised folic acid should change anyone’s treatment.

Read labels, if you wish to. In the UK, flour will list folic acid after the word flour. Wholemeal flour is exempt, flour from the smallest mills may be too, and the rule covers common wheat only.

What We Took Out, What We Put Back

I began with a bag of flour and the short list of vitamins printed on it, the nutrients that were removed when the grain was refined and then added back. After looking closely, I do not think the answer is as simple as either side makes it.

Folic acid does lower the risk of neural tube defects, and that deserves to be said plainly. But the benefit is concentrated in a small group, women who are or may become pregnant, while the exposure is shared by everyone. The molecule is synthetic, the human enzyme that handles it is slow in all of us, and unmetabolised folic acid appears in the blood. Real questions remain about the brain, about vitamin B12 and about vulnerable groups who were never specifically studied, and a population measure has not yet answered them. The claim that people with MTHFR variants cannot handle folic acid goes further than the science does, and so does the claim that nobody has anything to worry about.

What strikes me most is how we arrived here. We took the nutrients out of the grain for convenience, put a few of them back as additives, and spent the better part of a century working out which ones we had forgotten. The honest ask of the policy is not to stop, but to measure: to watch what happens to older people, to kidney patients, to anyone whose body handles folate differently, and to be willing to change course if the data demand it. It should also test the obvious next question: whether fortifying with the active form, 5-MTHF, could give the benefit without the leftovers. So far it has been tested on blood levels and not on birth defects, and it deserves to be. And at the table, the oldest principle still stands. The more of our nutrients that come with the food, instead of being added to it, the less we have to ask of a laboratory.

Questions readers ask