Food vs. Supplements: Why the Pigment That Colors Carrots Behaves Differently in a Pill

Beta-carotene from carrots, sweet potato, and leafy greens is protective — but high-dose supplements raised lung cancer risk in two landmark trials of smokers. The food-versus-pill story matters.

Beta-carotene is the orange pigment that colors carrots, sweet potato, pumpkin, and butternut squash, and the same molecule hidden inside the green of kale, spinach, and collards. Your body converts it into vitamin A on demand, only making as much as you need — which is why food sources are remarkably safe even at very high intake. The story turns when beta-carotene is isolated and concentrated into a pill: two landmark trials in the 1990s found that high-dose supplements actually increased lung cancer in smokers [1][2]. The lesson is foundational to how this site thinks about nutrition: a colorful plate of vegetables is almost always wiser than an isolated capsule.

What Beta-Carotene Is

Beta-carotene is one of about 600 carotenoids — fat-soluble plant pigments responsible for yellow, orange, and red colors in fruits and vegetables. It is the most abundant provitamin A carotenoid, meaning the body cleaves it into retinol (active vitamin A) using an enzyme called BCO1. The conversion is regulated: when vitamin A status is sufficient, the body slows the conversion and stores beta-carotene unchanged in fat tissue and skin. This negative feedback is why beta-carotene from food essentially cannot cause vitamin A toxicity, while preformed retinol (from liver or high-dose retinol supplements) can.

Beyond serving as a vitamin A precursor, beta-carotene functions as an antioxidant in its own right, quenching singlet oxygen and scavenging peroxyl radicals in lipid-rich environments where water-soluble antioxidants cannot reach.

Best dietary sources (beta-carotene per 100 g, approximate):

Sweet potato (cooked): 9,400 mcg
Carrot (raw): 8,300 mcg
Kale (cooked): 6,200 mcg
Spinach (cooked): 6,100 mcg
Butternut squash (cooked): 4,000 mcg
Pumpkin (cooked): 3,100 mcg
Cantaloupe: 2,000 mcg
Apricot (dried): 2,200 mcg
Romaine lettuce: 1,300 mcg
Red bell pepper: 1,600 mcg

A single medium sweet potato or one large carrot covers a full day of vitamin A needs, with massive headroom for storage and antioxidant use.

Why Cooking and Fat Matter

Beta-carotene is locked inside the rigid cell walls of plants. Raw carrots release only about 3-5% of their beta-carotene to digestion. Cooking, chopping, pureeing, and adding fat dramatically improve absorption:

Steam, boil, or roast vegetables to break cell walls
Pair with a fat source — olive oil, avocado, butter, full-fat yogurt
Mash or puree when possible (carrot soup, butternut puree, sweet potato mash)

A dressed salad delivers far more usable beta-carotene than dry leaves; carrots in soup release more than raw carrot sticks. The same logic that improves lycopene absorption applies to beta-carotene — these fat-soluble carotenoids need fat to enter the bloodstream.

The Famous Supplement Trials

Three large randomized trials in the 1990s tested isolated beta-carotene as a cancer-prevention supplement and produced unexpected results:

The ATBC trial in 29,133 Finnish male smokers gave 20 mg/day beta-carotene and found 18% more lung cancer in the supplemented group [1]
The CARET trial in 18,314 American smokers and asbestos workers gave 30 mg beta-carotene plus 25,000 IU vitamin A and was halted early after 28% more lung cancer and 17% more deaths in the supplemented arm [2]
The Physicians' Health Study in 22,071 men (mostly non-smokers) gave 50 mg every other day for 12 years and found no benefit and no harm [3]

The simplest interpretation: in heavy smokers, high concentrations of isolated beta-carotene appear to interact with the oxidative environment of damaged lung tissue to promote rather than prevent carcinogenesis. The effect has not been seen in non-smokers, has not been seen with food sources, and has not been seen with mixed-carotenoid intakes.

The practical guidance is now well-established: beta-carotene supplements above a few milligrams per day are not advised, especially for current or former smokers. The same nutrient from carrots, sweet potato, and leafy greens has consistently been associated with lower cancer risk in observational studies — including in smokers.

Genetic Variation in Conversion

Roughly 30-45% of adults carry common variants in the BCO1 gene that reduce conversion of beta-carotene to retinol by 30-70%. People with low-converter genotypes may need higher dietary beta-carotene (or some preformed vitamin A from animal foods like eggs, dairy, cod liver oil, or organ meats) to maintain optimal vitamin A status.

Practical Guidance

Eat one to two cups daily of beta-carotene-rich vegetables (carrot, sweet potato, kale, spinach, squash, pumpkin)
Cook them and add fat — butter on roasted carrots, olive oil on kale, coconut milk in pumpkin soup
Avoid isolated beta-carotene supplements above 3-5 mg/day, especially if you smoke or have ever smoked
If you take a multivitamin, prefer one with mixed carotenoids (alpha-carotene, beta-carotene, lutein, zeaxanthin, lycopene) at modest doses rather than high-dose beta-carotene alone
For age-related macular degeneration, the modern AREDS2 formula uses lutein and zeaxanthin instead of beta-carotene [5]

See our pages on vitamin A, lycopene, lutein and zeaxanthin, and astaxanthin for related carotenoids and how they fit together.

Evidence Review

The Smoker Harm Trials

The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study (ATBC) randomized 29,133 male Finnish smokers aged 50-69 to receive 20 mg/day beta-carotene, 50 mg/day alpha-tocopherol, both, or placebo for an average of 6.1 years [1]. The trial was designed to test whether antioxidant vitamins could prevent lung cancer. Instead, the beta-carotene group showed an 18% higher incidence of lung cancer (relative risk 1.18, 95% CI 1.03-1.36) and an 8% higher overall mortality. Subgroup analysis indicated the harm was concentrated in heavier smokers (those smoking 20+ cigarettes daily) and in men who consumed more alcohol.

The Carotene and Retinol Efficacy Trial (CARET) randomized 18,314 men and women at high lung-cancer risk — current/former smokers and asbestos-exposed workers — to 30 mg beta-carotene plus 25,000 IU retinol daily versus placebo [2]. After an interim analysis showed 28% more lung cancer (relative risk 1.28, 95% CI 1.04-1.57) and 17% higher all-cause mortality (relative risk 1.17, 95% CI 1.03-1.33) in the active arm, the trial was terminated 21 months early.

By contrast, the Physicians' Health Study randomized 22,071 male physicians to 50 mg beta-carotene every other day or placebo for 12 years [3]. Only 11% were current smokers and 39% former smokers. After 12 years, there was no difference in cancer incidence (relative risk 0.98, 95% CI 0.91-1.06), no difference in cardiovascular events, and no difference in all-cause mortality. The lack of harm in this lower-smoking cohort, combined with the harm in the high-smoking cohorts of ATBC and CARET, suggests the adverse interaction is specific to the oxidative environment of actively damaged lung tissue.

A 2010 meta-analysis by Druesne-Pecollo et al. pooled data from 13 RCTs encompassing over 200,000 participants [8]. Pooled beta-carotene supplementation (typically 20-50 mg/day) was associated with a small but statistically significant increase in overall cancer incidence (relative risk 1.10, 95% CI 1.03-1.17) and a non-significant trend toward higher mortality. Effects were most pronounced for lung and stomach cancers, especially among smokers and asbestos-exposed workers.

Strength of evidence for supplement harm in smokers: Strong. Three large RCTs converge on the same finding; doses tested (20-50 mg/day) are 3-15 times typical dietary intake.

The Food-Source Discrepancy

The supplement trials stand in stark contrast to observational data on dietary beta-carotene. Pre-trial cohort studies — the very evidence that motivated ATBC and CARET — had consistently shown that people eating more beta-carotene-rich foods had 30-50% lower lung cancer risk, even within smoker subgroups. After the trials, this association did not disappear from the observational literature; it has been replicated in numerous cohorts since.

Several explanations are now widely accepted:

Food sources deliver beta-carotene alongside hundreds of other carotenoids and polyphenols (alpha-carotene, lutein, zeaxanthin, lycopene, beta-cryptoxanthin, flavonoids, vitamin C, fiber) that may work synergistically and balance each other's pro-oxidant tendencies
Plasma concentrations from food are far lower than from 20-50 mg/day supplements; the dose-response curve appears non-monotonic
In hypoxic, oxidatively stressed tissue (smoker lungs), high beta-carotene concentrations may shift from antioxidant to pro-oxidant behavior, generating reactive oxidation products that damage DNA
Preformed vitamin A in CARET likely amplified the effect, since high circulating retinol can interfere with beta-carotene metabolism

This is the cleanest example in modern nutrition science of the principle that a nutrient extracted, concentrated, and administered in pharmacologic doses is not equivalent to the same nutrient embedded in a whole food matrix.

Macular Degeneration: From AREDS to AREDS2

The original Age-Related Eye Disease Study (AREDS, 2001) randomized 3,640 participants with intermediate or advanced AMD to a daily formula containing 500 mg vitamin C, 400 IU vitamin E, 15 mg beta-carotene, 80 mg zinc, and 2 mg copper [4]. Over 6.3 years, the formula reduced progression to advanced AMD by 25% (odds ratio 0.72, 99% CI 0.52-0.98) and reduced moderate vision loss by 19%. The protective effect was confirmed and the AREDS formula became standard ophthalmologic recommendation for intermediate AMD.

However, the 15 mg beta-carotene was problematic for current and former smokers. AREDS2 (2013) tested whether substituting 10 mg lutein + 2 mg zeaxanthin for beta-carotene maintained efficacy [5]. The trial randomized 4,203 participants. The lutein/zeaxanthin substitution showed equivalent or slightly better protection against AMD progression (hazard ratio 0.82, 95% CI 0.69-0.96 for lutein/zeaxanthin vs. beta-carotene, in the secondary analysis among the lowest dietary carotenoid intake quintile). Critically, lung cancer incidence in former smokers was 0.57% with lutein/zeaxanthin versus 2.0% with beta-carotene (P = 0.04). The modern AREDS2 formula now uses lutein/zeaxanthin and is recommended over beta-carotene for AMD prevention.

Strength of evidence for AMD: Strong for the AREDS-type formula in established intermediate/advanced AMD. The shift to lutein/zeaxanthin in AREDS2 represents a clear improvement on safety without losing efficacy.

Skin Photoprotection

A 2008 meta-analysis by Köpcke and Krutmann pooled seven controlled trials examining beta-carotene supplementation and minimal erythema dose (MED), the threshold UV exposure that produces sunburn [7]. Across studies, beta-carotene supplementation provided modest but statistically significant photoprotection. The protective effect was dose- and duration-dependent: studies of less than 10 weeks duration showed minimal benefit, while studies of 12+ weeks at 24-30 mg/day produced meaningful increases in MED. The authors estimated each additional month of supplementation increased MED by approximately 0.5 standard error units.

A foundational review by Stahl and Sies summarized the mechanistic basis [6]. Beta-carotene and other carotenoids accumulate in skin, where they quench singlet oxygen and triplet-state photosensitizers generated by UV exposure, reducing lipid peroxidation and downstream inflammatory signaling. They also suppress matrix metalloproteinase-1 (MMP-1) induction, the same collagen-degrading enzyme implicated in photoaging. Stahl and Sies emphasized that nutritional photoprotection is additive — not a replacement — for topical sunscreen and sun-avoidance behavior.

A separate dietary observation noted that consistent consumers of orange and dark-green vegetables develop measurable skin yellowing (carotenoid accumulation in stratum corneum) that correlates with both photoprotection and perceived health/attractiveness in cross-cultural studies.

Strength of evidence for photoprotection: Moderate. The effect is real but modest (approximately 1 minimal erythema dose increase after 12+ weeks of supplementation) and not a substitute for sunscreen. Dietary carotenoids from food appear to provide similar benefit at lower exposure levels.

Safety, Lycopenemia-Like Carotenodermia, and Drug Interactions

High dietary intake of beta-carotene-rich foods (e.g., daily juicing, repeated sweet-potato meals) can cause carotenodermia — a harmless yellow-orange tinting of palms, soles, and nasolabial folds. It is reversible within weeks of reducing intake and is a useful biomarker of saturated tissue stores rather than a sign of toxicity. It does not affect the sclera (unlike jaundice), which helps distinguish it clinically.

Beta-carotene supplements may interact with statins (reducing some lipid effects), bile-acid sequestrants (reducing absorption), orlistat and other fat blockers (reducing absorption), and may amplify alcohol-related hepatotoxicity at very high doses. Pregnant women should obtain vitamin A primarily from beta-carotene rather than preformed retinol, since beta-carotene cannot reach teratogenic vitamin A levels through dietary intake.

Bottom Line

The beta-carotene story is among the most pedagogically important in nutrition. Three converging lines of evidence support the same conclusion:

Diets high in carotenoid-rich vegetables are protective across nearly every studied outcome
Isolated, high-dose beta-carotene supplements are at best neutral and at worst harmful in smokers
Mixed dietary carotenoids and modern formulas (lutein/zeaxanthin) preserve benefit while removing the risk

Eat the carrots and sweet potatoes. Cook them with fat. Skip the high-dose pills.

References

The Effect of Vitamin E and Beta Carotene on the Incidence of Lung Cancer and Other Cancers in Male SmokersThe Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. New England Journal of Medicine, 1994. PubMed 8127329 →
Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular diseaseOmenn GS, Goodman GE, Thornquist MD, Balmes J, Cullen MR, Glass A, Keogh JP, Meyskens FL, Valanis B, Williams JH, Barnhart S, Hammar S. New England Journal of Medicine, 1996. PubMed 8602180 →
Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular diseaseHennekens CH, Buring JE, Manson JE, Stampfer M, Rosner B, Cook NR, Belanger C, LaMotte F, Gaziano JM, Ridker PM, Willett W, Peto R. New England Journal of Medicine, 1996. PubMed 8602181 →
A randomized, placebo-controlled, clinical trial of high-dose supplementation with vitamins C and E, beta carotene, and zinc for age-related macular degeneration and vision loss: AREDS report no. 8Age-Related Eye Disease Study Research Group. Archives of Ophthalmology, 2001. PubMed 11594942 →
Lutein + zeaxanthin and omega-3 fatty acids for age-related macular degeneration: the Age-Related Eye Disease Study 2 (AREDS2) randomized clinical trialAge-Related Eye Disease Study 2 (AREDS2) Research Group. JAMA, 2013. PubMed 23644932 →
Carotenoids and flavonoids contribute to nutritional protection against skin damage from sunlightStahl W, Sies H. Molecular Biotechnology, 2007. PubMed 17854613 →
Protection from sunburn with beta-Carotene — a meta-analysisKöpcke W, Krutmann J. Photochemistry and Photobiology, 2008. PubMed 18086242 →
Beta-carotene supplementation and cancer risk: a systematic review and metaanalysis of randomized controlled trialsDruesne-Pecollo N, Latino-Martel P, Norat T, Barrandon E, Bertrais S, Galan P, Hercberg S. International Journal of Cancer, 2010. PubMed 20232415 →