Case Study 2 — Folic Acid: What It Looks Like When Nutrition Science Wins

A public health and evidence case. The trials, the fortification programmes, and the outcomes are real.

Setup

This book has spent twelve chapters being sceptical, and a reader could reasonably have concluded by now that nutrition science mostly produces overturned claims and expensive urine.

So here is the other thing it produces, and it is worth a whole case study because almost nobody knows the story.

Folic acid and neural tube defects. It is nutrition's clearest success, it involved randomized trials rather than cohorts, it survived contact with policy, it has prevented an enormous amount of irreversible disability, and it is almost never taught as a triumph — because prevented birth defects have no constituency and no photographs.

What a neural tube defect is

The neural tube is the embryonic structure that becomes the brain and spinal cord. It closes by around the fourth week after conception — often before a person knows they are pregnant.

If it fails to close properly, the result is a neural tube defect:

  • Spina bifida — incomplete closure of the spine, ranging from mild to causing paralysis, bowel and bladder dysfunction, and hydrocephalus requiring lifelong management
  • Anencephaly — failure of the skull and brain to develop; not compatible with survival

These are not minor. They were, historically, among the most common serious congenital malformations — and in the mid-twentieth century, in some populations, rates were strikingly high.

Stage 1: the observation

From the 1960s onward, researchers noticed that neural tube defects clustered — geographically, by season, and by socioeconomic status. Poorer populations had higher rates. Rates varied between regions in ways that didn't look genetic.

Diet was an obvious candidate, and folate — required for DNA synthesis during rapid cell division, which is exactly what a closing neural tube is doing — was the obvious nutrient.

At this stage the evidence was observational, with all of Chapter 2's problems. Poorer populations differ in a hundred ways. This could easily have been another beta-carotene.

Stage 2: the trial

And this is where the story diverges from every other one in this book. Somebody ran the trial.

The MRC Vitamin Study, published in 1991, was a multicentre randomized controlled trial. It enrolled women who had previously had a pregnancy affected by a neural tube defect — a group at substantially elevated risk of recurrence — and randomized them to folic acid supplementation or not, in a factorial design with other vitamins.

The result was decisive: folic acid supplementation substantially reduced the recurrence of neural tube defects.

The effect was large enough, and clear enough, that the trial was stopped.

A separate randomized trial in Hungary, in women without a previous affected pregnancy, subsequently found a reduction in first occurrence — extending the finding from recurrence prevention to primary prevention.

📉 Evidence quality: rung 6. Randomized, controlled, with a hard clinical endpoint. In nutrition, this is close to as good as it gets — and note the contrast with every other section of Chapter 13, where we're arguing about cohorts.

Stage 3: the implementation problem

Here is where it gets interesting as a policy case rather than an evidence case.

The trial finding was unambiguous. The recommendation followed immediately: women who could become pregnant should take 400 µg of folic acid daily, starting before conception.

And it didn't work well enough.

The reason is the timing problem from §13.6: the neural tube closes by around week four. A substantial proportion of pregnancies are unplanned, and even planned ones are frequently recognized after the critical window has passed.

You cannot solve a problem that requires action before a decision, by advising people to act after the decision. Supplementation campaigns raised uptake, and uptake remained far from universal — and concentrated, predictably, among more educated and more affluent women, which is precisely the opposite of where the risk was highest.

Stage 4: fortification

So several countries did something more radical: they put folic acid in the flour.

Mandatory fortification of enriched grain products was introduced in the United States in the late 1990s, in Canada around the same time, and subsequently in a substantial number of other countries.

This is mass medication of a whole population without individual consent, applied to prevent a condition in a small subset. It is genuinely one of the more aggressive things public health has done, and it was contested.

The outcome: following fortification, countries that adopted it observed measurable declines in neural tube defect rates — reported in the range of roughly 20–35% in various national analyses, with the largest effects where baseline intake was lowest.

And crucially, the benefit reached the people advice hadn't. Fortification doesn't require planning, literacy, income, or a pharmacy. It works on unplanned pregnancies. It is one of the few public health interventions that reduces rather than widens health inequality, because it operates below the level of individual decision-making.

What this case demonstrates

1. Nutrition science can produce definitive answers. It requires a nutrient with a specific function, a defined outcome, a susceptible population, and someone willing to fund a randomized trial. Those conditions are rare — which is why most of this book is about cohorts — but they are not impossible.

2. Evidence and implementation are different problems. The MRC trial settled the science in 1991. It took population-level fortification to actually capture most of the benefit, because the recommendation required action in a window most people can't act in.

3. The intervention that worked required no behaviour change. Compare this to every other recommendation in this book — eat more fiber, drink less, hit your protein target — all of which depend on sustained individual action (Chapter 10's adherence problem). Fortification is what an intervention looks like when it doesn't depend on anyone doing anything.

4. And it has genuine costs worth stating. Fortification is imposed without consent. There has been legitimate debate about whether high folic acid intake could mask the anemia of B12 deficiency while neurological damage progresses — which is a real concern, particularly in older adults, and is an argument for checking B12 rather than against fortification. There has also been debate about unmetabolized folic acid in circulation, where the evidence remains inconclusive. The programme is a net good with real trade-offs, and pretending otherwise would be exactly the overconfidence this book criticizes.

💡 Aha moment. Notice what this case has that the beta-carotene story lacks, because the contrast is the lesson.

Both started with an observational association between a nutrient and a disease. Beta-carotene: inferred to a compound, tested as a high-dose pill in a replete population, harmed. Folate: tested at a physiological dose in a population with a specific elevated requirement, at the specific time the nutrient is needed, against a specific outcome the nutrient mechanistically influences.

Same starting point. Opposite result. And the difference is not luck — it's that the folate hypothesis specified who, when, why, and how much, and the antioxidant hypothesis specified more.

That distinction — correcting a deficiency in a susceptible group versus supplementing a replete one — is §13.6's shape, and it is the single most useful predictor of whether a supplement trial will work.

Discussion Questions

  1. Fortification is mass medication without individual consent. Construct the strongest objection. Then the strongest defence. Does the fact that it reduces health inequality change the ethical calculus?

  2. The recommendation to supplement preconceptionally was correct and insufficient. Name two other health recommendations that fail for the same structural reason — requiring action before a decision people haven't made yet.

  3. The masking-of-B12-deficiency concern is real. How should a policy weigh a definite benefit in one population against an uncertain harm in another? Who decides?

  4. Compare the folate story with the beta-carotene story point by point. Write the general rule that distinguishes them, in one sentence, and test it against two other supplements in this book.

  5. This is described as "almost never taught as a triumph" because prevented birth defects have no constituency. What follows for how public health communicates its successes? Is there a fix?

Your Turn

Find another nutrition intervention that genuinely worked — one where the evidence is strong, the implementation happened, and the outcome improved.

Candidates: iodized salt and goitre/cretinism · vitamin D fortification and rickets · niacin fortification and pellagra · vitamin K at birth and haemorrhagic disease of the newborn · oral rehydration solution and diarrhoeal mortality.

Pick one and trace it:

  1. What was the disease burden before? Find actual numbers.
  2. What was the evidence, and at what rung?
  3. What was the implementation — advice, supplementation, or fortification?
  4. What opposition was there?
  5. What did it prevent?

Then answer the question this case study is really about: had you heard of it?

Most people haven't heard of any of them, which tells you something about which nutrition stories propagate — and it is the same finding as Chapter 11's "zero for seven" and Chapter 12's asymmetric correction. Successes are invisible because prevented disease looks like nothing at all.