What Vitamin A Actually Is

Vitamin A is not one molecule. It is a family of fat-soluble compounds that share the ability to activate retinoid receptors in your cells, and the way your body gets them differs depending on what you eat. From animal sources, you get preformed vitamin A (retinol and retinyl esters), which is absorbed efficiently and stored directly in the liver. From plant sources, you get provitamin A carotenoids, primarily beta-carotene, which your body must convert into retinol before it can use them. This conversion is not trivial, and the gap between “eating beta-carotene” and “having usable vitamin A” is where most of the confusion lives.

The standard way to account for this difference is the retinol activity equivalent (RAE), which the Institute of Medicine adopted in 2001.¹ Under this system, 1 mcg RAE equals 1 mcg of retinol, 2 mcg of supplemental beta-carotene in oil, or 12 mcg of dietary beta-carotene from food. That 12:1 ratio for food-sourced beta-carotene replaced the old 6:1 estimate, which had overestimated how much vitamin A people were actually getting from vegetables by roughly double. The European Food Safety Authority still uses the older, more generous conversion factors, which creates some confusion if you’re comparing international recommendations.²

The Recommended Dietary Allowance is 900 mcg RAE per day for adult men and 700 mcg RAE per day for adult women.³ The Tolerable Upper Intake Level for preformed vitamin A (not beta-carotene) is 3,000 mcg RAE per day, set based on the risk of liver abnormalities and teratogenic effects. That distinction between preformed and provitamin A matters enormously for understanding toxicity, which we’ll get to.

A necessary disclaimer: Vitamin A metabolism intersects with conditions including liver disease, pregnancy, measles, and malabsorption disorders. If you have any of these, or suspect a deficiency, work with a medical provider. This article explains mechanisms and evidence. It is not a substitute for clinical evaluation.

How beta-carotene becomes vitamin A

When you eat a carrot or a sweet potato, the beta-carotene locked in the plant’s cellular matrix has to survive a gauntlet before it does anything useful. First, it has to be physically freed from the food matrix — the cell walls and protein structures of the plant. Then it needs to be incorporated into lipid micelles in the small intestine, which requires the presence of dietary fat and bile acids. Only after all of that does it reach the enterocytes (intestinal absorptive cells) where conversion can begin.⁴

At the enterocyte cell wall, beta-carotene is taken up by a membrane transporter called scavenger receptor class B, type 1 (SCARB1). Once inside, the enzyme beta-carotene 15,15′-oxygenase (encoded by the BCO1 gene) cleaves the beta-carotene molecule at its center, producing two molecules of retinal — the aldehyde form of vitamin A. Retinal is then reduced to retinol, bound to retinol-binding protein 2, esterified, and packed into chylomicrons for transport to the liver.

The system has a built-in feedback loop. When plasma retinol levels are adequate, both SCARB1 and BCO1 are downregulated, meaning your body absorbs less beta-carotene and converts less of what it does absorb.⁵ This is why beta-carotene supplementation doesn’t cause vitamin A toxicity even at very high doses: the body throttles the pipeline. At most, you’ll turn a bit orange from carotenodermia, which is harmless and reversible. This feedback mechanism does not exist for preformed vitamin A from animal sources or retinol supplements, which is why overdosing on retinol is possible and overdosing on beta-carotene effectively isn’t.

The conversion efficiency problem

Saying “12 mcg of dietary beta-carotene equals 1 mcg of retinol” is a useful shorthand, but it hides enormous individual variation. The actual conversion ratio measured in human studies ranges from about 3.8:1 to 28:1 by weight, depending on the food source, how it was prepared, the dose, and who is eating it.⁶

The food matrix is the biggest variable. Beta-carotene in carrots exists in a crystalline form within chromoplasts, which is relatively easy to disrupt through cooking and chewing. Spinach beta-carotene sits in chloroplast pigment-protein complexes that are harder to break down. In stable isotope studies, carrot beta-carotene showed a conversion factor around 15:1 while spinach came in around 21:1, despite both being pureed and cooked.⁷ Raw, whole-leaf spinach would be even worse. Beta-carotene delivered in oil capsules, by contrast, can achieve ratios as favorable as 3.8:1, which is why the RAE system distinguishes between supplemental and dietary sources.

Cooking, blending, and adding fat all improve bioavailability. A minimum of about 2.4 grams of fat per meal appears necessary to facilitate meaningful carotenoid absorption, though more is better within reason.⁸ Certain fibers also matter: pectin, guar, and alginates reduced carotenoid absorption by about 33-43% in one study, while wheat bran and cellulose had no effect.⁹

Genetics: the BCO1 variable

Your genes determine how efficiently your BCO1 enzyme works, and the variation here is striking. Two common single-nucleotide polymorphisms in the BCO1 coding region (R267S, rs12934922 and A379V, rs7501331) reduce the enzyme’s catalytic activity. Carriers of the double-variant allele showed a 57% reduction in beta-carotene conversion in vitro, with corresponding drops of 32-69% in the retinyl palmitate-to-beta-carotene ratio in blood after a test dose.¹⁰

Additional upstream polymorphisms reduce conversion efficiency by 48-59%, and the affected allele frequencies vary significantly between ethnic groups, ranging from 19% to 100% depending on the specific variant and population studied.¹¹ Up to 45% of the general population may qualify as “low responders” to dietary beta-carotene. These are people whose bodies are genuinely poor at converting plant-sourced carotenoids into usable retinol.

This has practical implications. Someone who eats a vegetable-heavy diet and assumes their beta-carotene intake covers their vitamin A needs might be meaningfully deficient if they carry BCO1 variants that cut their conversion rate in half. There is no standard clinical test for BCO1 genotype, but a serum retinol test can reveal whether your vitamin A status is adequate regardless of how you’re getting it.

What vitamin A does once you have it

Vitamin A’s best-known role is in vision. Retinal combines with the protein opsin in your retina to form rhodopsin, the photopigment required for low-light (scotopic) vision. This is why night blindness is the earliest clinical sign of vitamin A deficiency — the rod cells literally cannot regenerate their light-sensing pigment without a steady retinal supply.

But vision accounts for a relatively small fraction of the body’s vitamin A use. The more systemic functions are mediated primarily by retinoic acid, the oxidized form of vitamin A that acts as a signaling molecule by binding to nuclear receptors (RARs and RXRs) and regulating gene transcription.

Immune regulation

Retinoic acid is one of the primary regulators of mucosal immunity, the immune response that operates at barrier surfaces like the gut lining, respiratory tract, and skin. It does this through several mechanisms. Dendritic cells in the gut-associated lymphoid tissue metabolize retinol into retinoic acid during antigen presentation, and this retinoic acid instructs activated T cells to express gut-homing receptors (CCR9 and α4β7 integrin). Those receptors act as molecular zip codes, sending the T cells to mucosal surfaces where pathogens are most likely to enter.¹²

Retinoic acid also promotes the differentiation of Foxp3+ regulatory T cells (Tregs), which suppress excessive inflammatory responses and help maintain immune tolerance — the ability to distinguish between genuine threats and harmless substances like food proteins.¹³ In vitamin A deficiency, this balance shifts. Inflammatory T cell populations (Th1 and Th17 cells) become disproportionately active while regulatory responses weaken. The result is an immune system that handles infections poorly and simultaneously overreacts at mucosal surfaces.

This is why vitamin A supplementation in populations with high deficiency rates sharply reduces childhood mortality from measles and diarrheal diseases. The mechanism is specific: restoring the regulatory architecture that prevents immune responses from being simultaneously too weak against pathogens and too aggressive against the body’s own tissues.

Cell differentiation and tissue integrity

Retinoic acid regulates the differentiation of epithelial cells throughout the body, from skin and cornea to the respiratory tract lining. Without adequate vitamin A, epithelial tissues undergo squamous metaplasia — they lose their normal specialized structure and become thick, dry, and keratinized. In the eye, this progression from night blindness to corneal drying (xerophthalmia) to permanent corneal scarring and blindness remains one of the leading causes of preventable blindness worldwide, affecting an estimated one-third of children under five in developing countries.¹⁴

Vitamin A is also essential for embryonic development, where retinoic acid acts as a morphogen (a signaling molecule that tells cells where they are in the developing body and what type they should become). This is the same mechanism that makes excessive preformed vitamin A teratogenic, because too much retinoic acid disrupts the precise concentration gradients that guide normal organ formation.

Toxicity: why preformed vitamin A is the concern

Hypervitaminosis A is a risk exclusively from preformed vitamin A (retinol and retinyl esters). Because carotenoid conversion is feedback-regulated, excess beta-carotene accumulates as carotenoids in subcutaneous fat, turning your skin yellow-orange, rather than being forced through the conversion pipeline into retinol. No cases of hypervitaminosis A from beta-carotene have been documented.

Preformed vitamin A, by contrast, is absorbed at 70-90% efficiency via passive diffusion in the intestine, with no feedback mechanism to slow absorption when stores are full.¹⁵ Acute toxicity can occur from single doses above about 150,000 mcg (500,000 IU), with symptoms including nausea, vomiting, headache, and blurred vision. Chronic toxicity generally develops at sustained intakes well above the 3,000 mcg RAE upper limit, with clinical signs (dry skin, joint pain, fatigue, liver damage, bone loss) typically reported at long-term intakes in the range of 8,000-10,000 mcg RAE per day or higher. The UL itself is set conservatively, at roughly three times the RDA, to provide a safety margin before those clinical thresholds.

The teratogenic risk deserves particular attention. In a large prospective study of over 22,000 pregnancies, women taking more than 10,000 IU of preformed vitamin A per day from supplements had a roughly 4.8-fold increase in cranial neural crest defects in their babies, with an apparent threshold near 10,000 IU per day. The increased risk was concentrated among women who consumed high levels before the seventh week of gestation.¹⁶ This risk applies exclusively to preformed vitamin A; animal and human data consistently show no teratogenicity from beta-carotene at any studied dose.

This asymmetry is the primary reason many supplement manufacturers have shifted toward beta-carotene as their vitamin A source, and it’s worth understanding when reading labels. A multivitamin listing 900 mcg RAE of vitamin A could be delivering that as retinyl palmitate, as beta-carotene, or as a mix. The distinction matters for both the effectiveness of the supplement (especially if you’re a poor beta-carotene converter) and the safety profile (especially if you’re pregnant or could become pregnant).

The ATBC and CARET trials: when beta-carotene caused harm

If the feedback regulation of beta-carotene conversion makes toxicity essentially impossible, how did beta-carotene supplementation increase lung cancer rates in two major clinical trials?

The Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) Study, published in 1994, randomized 29,133 male smokers in Finland to receive either 20 mg of beta-carotene daily, 50 mg of alpha-tocopherol, both, or a placebo for 5-8 years. The beta-carotene arm showed an 18% increase in lung cancer incidence and an 8% increase in overall mortality.¹⁷

The Beta-Carotene and Retinol Efficacy Trial (CARET) tested 30 mg of beta-carotene plus 25,000 IU of retinyl palmitate daily in over 18,000 current and former heavy smokers plus asbestos-exposed workers. This trial was stopped early when the intervention group showed a 28% increase in lung cancer and 17% increase in overall mortality.¹⁸

A meta-analysis of the combined trial data showed that among current smokers specifically, beta-carotene supplementation was associated with a 24% increase in lung cancer risk. Among former smokers and never-smokers, there was no statistically significant increase.¹⁹

Meanwhile, the Physicians’ Health Study — which gave 50 mg of beta-carotene every other day to nearly 22,000 male physicians, 89% of whom were non-smokers — found no significant effect on cancer or cardiovascular disease after 12 years.

What probably happened

The leading hypothesis centers on what happens when high concentrations of beta-carotene meet the oxidative environment of a smoker’s lungs. Under normal conditions, BCO1 cleaves beta-carotene symmetrically at its central double bond to produce retinal. But in the presence of cigarette smoke oxidants, beta-carotene undergoes asymmetric, non-enzymatic oxidative cleavage that produces reactive apo-carotenoids and beta-carotene epoxides rather than clean retinal.²⁰ These cleavage products can interfere with retinoid signaling pathways in lung tissue and may promote cell proliferation rather than the orderly differentiation that retinoic acid normally supports.

Animal studies in ferrets (which metabolize carotenoids similarly to humans) confirmed that high-dose beta-carotene combined with smoke exposure produced increases in lung cell proliferation markers, while the same dose of beta-carotene without smoke exposure did not.²¹ The presence of antioxidants like alpha-tocopherol shifted the cleavage back toward the normal central pathway, suggesting that the pro-oxidant environment of the smoke-exposed lung is the critical variable.

The doses in these trials were also far above anything achievable through diet. Twenty to 30 mg of supplemental beta-carotene per day is equivalent to roughly five medium-sized carrots’ worth of beta-carotene delivered in a bioavailable oil capsule, every day, for years, in people whose lung tissue was already under severe oxidative stress. Beta-carotene from food remains safe at any plausible dietary intake. The danger is specific to pharmacological doses of isolated supplements taken by current smokers, and probably by recent former smokers as well.

Beta-carotene as an antioxidant (independent of vitamin A conversion)

Not all beta-carotene that enters your body gets converted to retinol. Intact beta-carotene circulates in the blood and accumulates in tissues, where it can function as an antioxidant independent of vitamin A activity. Its conjugated polyene chain (the same structural feature that gives it its orange color) makes it an efficient quencher of singlet oxygen and a scavenger of peroxyl radicals.

This antioxidant activity is likely what drove the observational epidemiology that prompted the ATBC and CARET trials in the first place. Dozens of cohort and case-control studies consistently found that people with higher blood beta-carotene levels and higher dietary intake of carotenoid-rich vegetables had lower rates of lung cancer and cardiovascular disease. The ATBC Study itself showed this pattern in its placebo group: participants in the lowest quartile of baseline serum beta-carotene had 33% higher lung cancer incidence than those in the highest quartile.²²

The disconnect between “higher serum beta-carotene correlates with lower cancer risk” and “supplemental beta-carotene increases cancer risk in smokers” illustrates a principle that applies broadly across nutrition science: a marker of healthy dietary patterns is not necessarily the active ingredient. People who eat more vegetables have higher beta-carotene levels, but they also consume hundreds of other bioactive compounds and tend to differ from low-vegetable consumers in ways that are difficult to fully adjust for in observational studies. Isolating one compound and delivering it at pharmacological doses to high-risk populations is a fundamentally different intervention than eating more broccoli.

Food sources and practical absorption

Beta-carotene is abundant in orange and yellow vegetables (carrots, sweet potatoes, butternut squash, pumpkin), dark leafy greens where the orange pigment is masked by chlorophyll (spinach, kale), and some fruits (cantaloupe, mangoes, apricots). Preformed vitamin A comes from liver (by far the richest source), dairy products, eggs, and fortified foods.

For people relying primarily on plant sources, a few preparation strategies meaningfully improve beta-carotene absorption. Cooking breaks down the cell walls that trap carotenoids. Pureeing or finely chopping does the same mechanically. Eating carotenoid-rich foods with some dietary fat — even a small amount — is necessary for micelle formation and absorption. A salad with no dressing delivers substantially less usable beta-carotene than the same salad with olive oil or avocado.

Even with optimal preparation, the 12:1 conversion ratio means vegetarians and vegans need to eat significantly more carotenoid-rich plant food than the old 6:1 estimate suggested to hit the RDA from plant sources alone. The IOM calculated that a strict vegetarian diet can meet the RDA, but only if it consistently includes deeply colored fruits and vegetables as a planned dietary focus, not as an afterthought.²³ For people with BCO1 polymorphisms that further reduce conversion efficiency, the gap becomes wider and may warrant either monitoring retinol status via blood testing or considering a preformed vitamin A supplement at a safe dose.

Supplementation: when it makes sense and what to watch for

Most people eating a Western diet that includes some animal products, fortified foods, or both are not at meaningful risk of vitamin A deficiency. The liver stores enough retinol to sustain normal function for months on a deficient diet. Average daily beta-carotene intake in the US and Europe falls in the 2-7 mg range from food alone.

Supplementation with beta-carotene (rather than preformed retinol) has a simpler safety profile because of the feedback regulation on absorption and conversion. The primary safety signal — the lung cancer risk in smokers from the ATBC and CARET trials — applies specifically to high-dose isolated beta-carotene supplements (20-30 mg/day) in current and recent former smokers. For non-smokers, the evidence from the Physicians’ Health Study and other trials shows no increased risk.

The people most likely to benefit from supplementation are those relying heavily on plant sources for vitamin A. Vegetarians and vegans who don’t know their BCO1 genotype are better served by a low-dose preformed retinol supplement (within the RDA, well under the 3,000 mcg RAE upper limit) than by a beta-carotene supplement, since retinol bypasses the conversion bottleneck entirely. Some combination supplements provide both.

Pregnancy adds a layer of complexity. Preformed vitamin A intake from all sources combined should stay below 3,000 mcg RAE (10,000 IU) per day, with particular caution in the first trimester when the teratogenic risk is concentrated. Beta-carotene poses no known teratogenic risk at any dose. Most modern prenatal formulations use beta-carotene rather than retinyl palmitate for exactly this reason, but it’s worth checking the label.

Smokers should avoid high-dose beta-carotene supplements entirely. Dietary carotenoid intake from food is not a concern here; no study has linked it with increased cancer risk. The problem is specific to pharmacological-dose supplements in the context of ongoing smoke exposure.

One connection worth noting: zinc deficiency impairs retinol-binding protein synthesis and can reduce the body’s ability to mobilize vitamin A from liver stores. Adequate zinc status supports normal vitamin A metabolism, which makes zinc one of those nutrients that rarely gets mentioned in vitamin A discussions but probably should be.

Conclusion

Most of the confusion around vitamin A stems from treating it as a single nutrient when it’s really two separate supply chains with different absorption rules, different safety profiles, and different failure modes. Preformed retinol is efficient but can accumulate to toxic levels. Beta-carotene is self-limiting but depends on a conversion enzyme that works poorly in a large fraction of the population. Understanding which form you’re getting, and how well your body handles it, matters more than hitting a number on a label.

The ATBC and CARET trials remain among the most important cautionary examples in supplement science. What they demonstrated was that pharmacological-dose supplementation of an isolated compound can behave completely differently than the same compound consumed as part of a varied diet. This lesson extends well beyond vitamin A and into the broader question of why eating vegetables consistently outperforms taking pills in nearly every population-level health outcome that’s been studied.

For most people in developed countries, vitamin A deficiency is unlikely with a reasonably varied diet. For those eating primarily plant-based diets, or for anyone with symptoms that could indicate deficiency (persistent night vision problems, dry skin, frequent infections), a serum retinol test is the straightforward way to know where you stand.

Generate Your Stack. Avoid Conflicts. Optimize Absorption.

Most supplement advice ignores timing, interactions, and dose-response curves. Smart Stack Builder doesn't. Get a personalized daily schedule with interaction flags and dosage guidance, built on the same evidence base as this article.

Build Your Personalized Stack — $9.99

What vitamin A actually is (and why the label is confusing)