Micronutrients & DRIs

TL;DR. Vitamins and minerals are the catalytic chemistry your cells run on — small amounts of specific molecules enzymes can't work without. The Dietary Reference Intake framework (RDA, AI, EAR, UL, and the 2017 CDRR) sets population targets, but the dose that prevents scurvy isn't the dose that optimizes long-term risk. The big surprise of the last twenty years: isolating these nutrients into pills mostly doesn't work. Vitamin D for fractures, fish oil for heart disease, antioxidants for cancer — most have failed in large RCTs. "Food first" is a real rule, not a slogan.

What you'll learn

• What the DRI framework is in plain English: RDA, AI, EAR, UL, and CDRR.
• Why "enough to prevent deficiency" and "enough to optimize chronic-disease risk" are different numbers.
• Four nutrient stories worth knowing for most adults: vitamin D, folate/B12, magnesium, iron.
• Why supplement RCTs keep failing — the 500,000-person Mendelian randomization of vitamin D and fractures, SELECT, the fish oil meta-analyses.
• When supplements actually make sense: folic acid pre-conception, B12 for vegans and older adults, vitamin D in northern winters, iron for diagnosed deficiency.
• The nutrient-network logic (one-carbon metabolism, hepcidin) — why pulling a chemical out of food rarely behaves like the food.

The DRI framework, without the acronym soup

Whenever a label or doctor quotes a daily target for a vitamin or mineral, the number almost always comes from the Dietary Reference Intakes, set by expert panels at the U.S. and Canadian National Academies. The DRIs replaced the older single-number RDAs in 1997. They are a family of five reference values, each answering a different question.

The Estimated Average Requirement (EAR) is the median — the intake meeting the needs of about half of healthy people in a given age and sex group. Statisticians use it to assess populations; individuals don't aim for it. The Recommended Dietary Allowance (RDA) is the EAR plus about two standard deviations — high enough to cover 97-98 percent of healthy people. The RDA is what you see on food labels. The Adequate Intake (AI) is a best-judgment value used when evidence isn't strong enough to set an EAR — calcium in some age groups, choline. The Tolerable Upper Intake Level (UL) is the ceiling — the highest chronic daily intake unlikely to cause harm. For fat-soluble vitamins (A, D, E, K) and several trace minerals, the UL matters more than people realize.

In 2017 the National Academies added a fifth category for chronic-disease endpoints: the Chronic Disease Risk Reduction Intake (CDRR). The first was set for sodium. CDRRs exist because the older machinery was built to prevent classical deficiencies (scurvy, beriberi, pellagra, rickets) and doesn't translate cleanly to slow-onset disease.

Practical implication: "100% Daily Value of vitamin C" on a label is an RDA-derived target for preventing deficiency, not an optimized target for chronic-disease prevention. DRIs are population values calibrated to age, sex, and life stage.

The deficiency-versus-optimal gap

For most of the twentieth century, nutrition science was organized around classical deficiency diseases — and it worked. Scurvy responded to vitamin C, beriberi to thiamine, pellagra to niacin, rickets to vitamin D, pernicious anemia to B12. These diseases largely vanished from industrialized countries once food supply, fortification, and supplementation closed the gaps. The EARs and RDAs ask: how much prevents the classical deficiency and supports short-term function?

The harder question is different: how much minimizes the risk of cardiovascular disease, cancer, dementia, and fracture? Those endpoints take decades, are multi-causal, and depend on dietary patterns rather than single nutrients. The intake that prevents scurvy is well under 100 mg of vitamin C; whether 200, 500, or 2,000 mg is "optimal" for chronic-disease prevention has no honest answer yet.

The framework acknowledges this. Chapter 109 of Modern Nutrition in Health and Disease notes that the methodology struggles with chronic-disease endpoints and that CDRRs are the first formal attempt to fix it. Takeaway: hitting the RDA means you're very unlikely to develop a classical deficiency. It doesn't mean you're at the "best" intake for long-term risk reduction, and more isn't always better.

The four nutrient stories that matter for most adults

Most adults eating a varied diet don't need to think about most of the thirty-plus essential micronutrients individually. But four are worth knowing by name.

Vitamin D. Skin makes it from UVB sunlight; liver and kidneys activate it; it functions less like a vitamin and more like a hormone, regulating calcium absorption and bone remodeling. People above roughly San Francisco's latitude can't make meaningful D in winter; older adults, darker skin, and indoor work compound the risk. Walter Willett recommends a daily multivitamin partly to cover ~1,000 IU of D. But the evidence for what supplemental D does turns out to be weaker than the case for the deficiency itself — more below.

Folate and B12. These run one-carbon metabolism: the network that methylates DNA, makes neurotransmitters, recycles homocysteine, and supports rapid cell division. Folate deficiency in early pregnancy causes neural tube defects (spina bifida, anencephaly); B12 deficiency causes megaloblastic anemia and progressive neurologic damage that, if missed, becomes permanent. Folate sits in leafy greens, beans, and (since 1998 in the U.S.) fortified grain. B12 lives almost exclusively in animal foods — the one nutrient where strict vegans must act. Older adults run additional B12 risk as stomach acid and absorption decline.

Magnesium. A cofactor in more than 300 enzymatic reactions — ATP production, muscle contraction, blood pressure regulation, insulin signaling. Most Americans fall short of the RDA. Severe deficiency is uncommon but suboptimal status correlates in cohorts with higher type 2 diabetes and cardiovascular disease. Sources: leafy greens, nuts and seeds, whole grains, legumes — exactly the foods Western diets short.

Iron. Iron carries oxygen in hemoglobin and runs the electron transport chain. Iron deficiency is the most common nutrient deficiency in the world. The two at-risk groups in industrialized countries are menstruating women and toddlers. Heme iron from animal foods absorbs at 15-35 percent; non-heme iron from plants at 2-20 percent depending on the meal. Absorption is regulated centrally by hepcidin and modulated meal-by-meal by vitamin C (boosts) and calcium and tea/coffee tannins (block). The textbook case for why "iron in the food" isn't "iron in you."

Why supplement RCTs mostly fail

If micronutrients are good for you, and the foods containing them are good for you, then supplementing them in pill form should also be good for you. That's the intuition. In randomized trials, it's mostly wrong.

Vitamin D and fractures. The case rested on the idea that higher 25-hydroxyvitamin D levels would lower fracture risk. Tim Spector — the Kings College epidemiologist who wrote thirty papers on vitamin D over twenty-five years — stopped taking it when the evidence turned. The most rigorous test came from a Mendelian randomization study using inherited genetic variants as a proxy for lifetime exposure, side-stepping observational confounding. The Trajanoska BMJ analysis covered ~500,000 people and 188,000 fractures. The effect of higher D on fracture risk was essentially zero. High-dose D has also been linked in trials to increased falls and fractures.

Fish oil and heart disease. VITAL (25,000 people) and a Cochrane meta-analysis of 79 trials (~112,000 people) both found no meaningful reduction in heart disease, stroke, or cardiovascular mortality from omega-3 supplementation. NICE dropped its recommendation; the American Heart Association reversed.

Antioxidants and cancer. SELECT randomized ~35,000 men to vitamin E, selenium, both, or placebo, expecting a reduction in prostate cancer. It found the opposite: vitamin E increased prostate cancer risk by about 17 percent, and the trial was stopped early. An earlier beta-carotene trial in smokers (CARET) was stopped because it increased lung cancer rates. Multivitamin pools of more than half a million users show slightly elevated cancer and heart-disease mortality, not the protection hoped for.

Why? The leading hypothesis is whole-food synergy. Pollan's image: a leaf of thyme contains dozens of antioxidant compounds acting in concert, embedded in a fiber matrix, modulated by other phytochemicals. Pull beta-carotene out as a yellow capsule and you've removed it from every interaction that gave it its effect in food. Whole tomatoes lower prostate cancer risk in cohort studies; isolated lycopene does not. The dose-response curve in food differs from the curve in pills, and the supplement industry has pretended otherwise for forty years.

Food first vs. supplement first

"Food first" is the operational principle of the registered-dietitian profession. Assume the default route to a nutrient is eating food that contains it. Use supplements as gap-fillers for specific populations under specific conditions, not as routine insurance. Anti-reflex, not anti-supplement.

Where supplements have a defensible case:

• Folic acid before and during early pregnancy. The strongest case. Folic acid fortification of U.S. flour starting in 1998 cut neural tube defects by ~70 percent — one of the great public health wins of the last thirty years. 400 mcg daily before conception, 600 mcg during. Waiting until pregnancy is confirmed is too late; the neural tube closes by week four.
• B12 in strict vegans and many older adults. B12 isn't made by plants. Vegans need a fortified food or supplement. Adults over sixty often have reduced stomach acid (atrophic gastritis) and absorb dietary B12 poorly; after fifty, get most B12 from fortified foods or supplements.
• Vitamin D in northern latitudes and at-risk groups. Above roughly San Francisco's latitude, no meaningful D synthesis in winter. The RCT evidence for what supplementation prevents is weak; the evidence that deficiency is real, common, and easy to correct is strong.
• Iron in pregnancy and in women with documented deficiency. Don't supplement iron without measuring. Menstruating and pregnant women should be screened.

Where supplements are at best wasteful and at worst harmful: high-dose preformed vitamin A (retinol) damages bones; high-dose vitamin E raises prostate cancer; high-dose beta-carotene raises lung cancer in smokers; routine calcium supplements may modestly raise heart-attack and stroke risk; mega-dose multivitamins have failed to outperform food in any chronic-disease endpoint. The global supplement market is heading toward $200 billion with virtually no FDA premarket review under the 1994 DSHEA. The category, as a category, is unregulated.

Nutrient interactions — the network you can't see on a label

A label tells you how much of each nutrient is in a serving. It cannot tell you how those nutrients behave inside a metabolic network. Two examples.

One-carbon metabolism. Folate, B12, B6, choline, riboflavin, and methionine work together to transfer single-carbon groups. The network methylates DNA, recycles homocysteine to methionine, synthesizes nucleotides, and makes neurotransmitters. A common variant in MTHFR (~10 percent homozygous) reduces folate-cycle efficiency. Inadequate B12 traps folate in a metabolically dead form. Choline deficiency stresses the network even when folate is adequate. None of this shows up in the folate column on a label — deficiencies present as failures of a whole network, and supplementing one nutrient may not fix it if another is missing.

Iron and hepcidin. Iron absorption isn't a function of how much iron you ate — it's a function of how much your body decides it needs, mediated by hepcidin. High body iron suppresses absorption; low promotes it. On top of that, the meal matters: vitamin C can roughly triple non-heme iron absorption; calcium halves it; tea/coffee tannins cut it further. A vegetarian who pairs lentils with bell peppers and a squeeze of lemon, then waits an hour for her tea, is doing something pharmacologically meaningful the milligrams column doesn't capture.

The unit of analysis isn't really the nutrient. It's the food, the meal, the diet, and the eater. DRIs are useful population scaffolding, not a recipe.

Frequently Asked Questions

Should I take a daily multivitamin?

A small effect either way. Willett recommends one as cheap insurance, especially for the D. Spector dropped his after the evidence shifted. The largest trials show essentially no mortality benefit and a small possible signal of harm. If you eat a plant-rich diet, a multivitamin won't do much. If you eat a narrow Western diet, fixing the diet is a better lever.

Why does the RDA differ between countries?

Different panels weigh the same evidence differently. UK, EU, and U.S. values usually land within a factor of two of each other.

Is more vitamin C useless above a certain dose?

Above about 200 mg a day, plasma vitamin C plateaus — the kidneys excrete the rest. Megadoses don't reduce colds or treat cancer in proper trials, despite Linus Pauling's advocacy.

What about vitamin D blood tests?

The USPSTF doesn't recommend routine screening because evidence that knowing your level and acting on it improves outcomes is weak. If you're in a high-risk group (older, darker skin, northern latitude, indoor work, osteoporosis), discuss testing with your clinician.

Are "whole-food" supplements better than synthetic?

Sometimes marginally, often not at all, usually more expensive. For most water-soluble vitamins the synthetic form is chemically identical and equally bioavailable.

What's the practical bottom line?

Eat a diverse, mostly plant-rich diet. Take folic acid if you might become pregnant. Take B12 if you're vegan or over sixty. Consider vitamin D in northern winters. Screen for iron if you have heavy periods or are pregnant. Otherwise save the money.

How do I know if I'm actually deficient?

Classical deficiency symptoms are well-described and your clinician can test for them. Most "deficiencies" diagnosed online are either not real or not actionable. The biggest U.S. shortfalls are fiber, potassium, magnesium, vitamin D, and calcium — four of which are best addressed by adding plants, not pills.

Sources

1. Yaktine, A. L. "Dietary Reference Intakes." Modern Nutrition in Health and Disease, 12th ed. (2026), Ch 109.
2. Tucker, K. L., et al., eds. Modern Nutrition in Health and Disease, 12th ed. (2026) — iron (Ch 10), vitamin D (Ch 18), one-carbon metabolism (Chs 26-27, 30), micronutrient deficiency strategies (Ch 115).
3. Willett, W. C. Eat, Drink, and Be Healthy — Ch 11, "Take a Multivitamin for Insurance."
4. Spector, T. Spoon-Fed (2020) — Ch 5, "The Supplements Really Don't Work."
5. Duyff, R. L. Complete Food and Nutrition Guide, 5th ed. (Academy of Nutrition and Dietetics).
6. Pollan, M. In Defense of Food (2008) — whole-food synergy.
7. Trajanoska, K., et al. "Assessment of the genetic and clinical determinants of fracture risk: genome-wide association and Mendelian randomisation study." BMJ 2018; 362: k3225.
8. Klein, E. A., et al. "Vitamin E and the risk of prostate cancer: SELECT." JAMA 2011; 306(14): 1549-56.
9. CDC. Folic acid fortification of the U.S. food supply, 1998.
10. National Academies. Dietary Reference Intakes for Sodium and Potassium (2019) — introduction of the CDRR.

Related modules

• ← C3: Macronutrients
• C5: Inside your cells →
• C7: You're not average

Related glossary terms

• Dietary Reference Intakes
• Recommended Dietary Allowance (RDA)
• Adequate Intake (AI)
• Tolerable Upper Intake Level (UL)
• Estimated Average Requirement (EAR)
• Chronic Disease Risk Reduction Intake (CDRR)
• Vitamin D
• Folate
• One-carbon metabolism
• Food first