The Antioxidant Paradox: Why Observational Studies and Randomized Trials Disagreed
Why people who eat antioxidant-rich foods seem healthier, yet large trials of antioxidant pills mostly found no benefit, and what that tells us about reading studies.
Content last updated . This page has not been medically reviewed.
Quick answer
The "antioxidant paradox" is a mismatch. Studies that watch what people eat find that those who eat more fruit and vegetables, which are rich in antioxidants, tend to have lower risks of chronic disease. [1] But when researchers gave purified antioxidants as pills in randomized trials, the results mostly did not show the same benefit, and some showed harm. [1][2] The two kinds of study are answering different questions, so they do not contradict each other as much as the headline suggests. This page is information only, not personal advice.
How to read the evidence labels
Labels used here are explained on Antioxidants Explained: mixed, no good evidence of benefit, evidence of harm, preclinical. A finding about food is not a finding about a supplement.What observational studies saw
Observational studies follow people who choose their own habits. The National Cancer Institute (NCI) says studies of antioxidant supplements and cancer in this style "have yielded mixed results", and warns that such studies cannot adequately control for biases that might influence outcomes. [3] The ATBC trial began partly because epidemiologic evidence linked high blood levels of vitamin E and beta-carotene with lower lung cancer risk. [4]What the trials found
| Trial or review | Who and what was tested | Result | Label |
|---|---|---|---|
| ATBC (Finland) | 29,133 male smokers; alpha-tocopherol 50 mg and/or beta-carotene 20 mg a day (study amounts) [4] | Higher lung cancer incidence with beta-carotene [4] | Evidence of harm (in smokers) |
| CARET (US) | 18,314 smokers, former smokers and asbestos-exposed workers; beta-carotene plus retinol [5] | Relative risk of lung cancer 1.28; stopped early [5] | Evidence of harm (in this group) |
| Heart Protection Study (UK) | 20,536 adults with vascular disease or diabetes; vitamin E, vitamin C, beta-carotene daily (study amounts) [6] | No significant reduction in 5-year vascular disease, cancer or other major outcomes; the regimen appeared safe [6] | No good evidence of benefit |
| HOPE and HOPE-TOO | People with vascular disease or diabetes; natural-source vitamin E 400 IU daily (study amount) [7] | No prevention of cancer or major cardiovascular events; possible higher heart failure risk [7] | No good evidence of benefit; possible harm |
| SELECT (US, Canada) | 35,533 men; vitamin E and/or selenium [8] | More prostate cancer with vitamin E alone [8] | Evidence of harm (this form and group) |
| Nine NCI-listed cancer-prevention trials | Various antioxidant combinations | No evidence of benefit in primary cancer prevention [3] | No good evidence of benefit |
Details of the harm signals are on the high-dose risks table, and the pooled reviews on the Cochrane reviews page.
Related
Do antioxidant supplements work? · Beta-carotene, smokers and the ATBC and CARET trialsFrequently asked questions
What is the antioxidant paradox?
It is the gap between observational findings, where antioxidant-rich diets go with lower disease risk, and trial findings, where antioxidant pills mostly showed no benefit. [1]
Does this mean fruit and vegetables do not help?
No. The NCCIH says there is extensive evidence that people who eat more fruit and vegetables have lower risks of chronic diseases. [1] The trials tested purified supplements, not foods.
What is confounding?
It is when something else that travels with the habit being studied, such as exercise or regular check-ups, explains part of the result. The NCCIH gives that example for fruit and vegetable eaters. [1]
Did any antioxidant trial show harm?
Yes. Beta-carotene raised lung cancer incidence in male smokers in ATBC and in the CARET group, and vitamin E alone was linked to more prostate cancer in SELECT. [4][5][8]
Should I take antioxidant supplements to be safe?
The sources do not support using them to prevent cancer or heart disease. [1] Ask a clinician or pharmacist before starting any supplement, especially with medicines or in pregnancy.
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
Why the two can disagree
| Possible reason | What the sources say |
|---|---|
| Other lifestyle differences (confounding) | The NCCIH notes that benefits of a high fruit and vegetable diet may be due to other aspects of people's lifestyles that go with eating these foods. In a US rural study, eating five or more servings daily went along with moderate physical activity and a recent routine medical exam. [1] |
| Other substances in food | The benefit may come from other substances in the foods, not the antioxidants alone. [1] |
| Purified chemical vs whole food | Antioxidants taken as purified chemicals might act differently from those in foods, which contain complex mixtures of antioxidants, vitamins and minerals. [1][3] |
| Amount | High doses in supplements may have different effects than the smaller amounts in foods. [1] |
| Antioxidants are not interchangeable | The NCCIH says different antioxidants have different properties and may not be interchangeable. [1] |
| Too much may interfere | Large amounts of antioxidants may interfere with normal cell defenses and signaling. [1] |
Randomized trials assign people to a supplement or a placebo by chance, so they lack most of the biases that limit observational studies. [3] That is why the NCI treats them as the strongest evidence on benefit or harm. [3]
Limits and uncertainty
- Trials tested particular compounds, amounts, forms and populations. They do not show that every antioxidant, at any amount, in any person, fails. NCCIH points to one exception: certain antioxidant-plus-zinc supplements slowed progression in people who already had intermediate age-related macular degeneration. [1] See AREDS and AREDS2 results.
- Trials in smokers do not transfer automatically to non-smokers. In the Physicians' Health Study I, beta-carotene had no effect on cancer or all-cause mortality in either smokers or non-smokers. [3]
- This page covers supplement-level evidence. Whether a diet rich in plant foods is protective is a food question: see antioxidant foods vs supplements.
- Proper evidence not found that a supplement reproduces the effect seen with whole diets; we looked in the NCI and NCCIH summaries and the trial reports above.
Safety and when to ask
"Natural" and "antioxidant" do not mean safe at any amount. Antioxidant supplements may interact with medicines and cancer treatment, and pregnancy and children need a clinician's advice. [1] Do not stop a prescribed medicine to use a supplement. Talk to a clinician or pharmacist, and see medication interactions and quality testing and certifications.
References
- Antioxidant Supplements: What You Need To Know. National Center for Complementary and Integrative Health (US), last updated July 2023. Read 2026-10-23.
- Bjelakovic G, et al. Antioxidant supplements for prevention of mortality in healthy participants and patients with various diseases. Cochrane Database Syst Rev 2012. Read 2026-10-23.
- Antioxidants and Cancer Prevention. National Cancer Institute (US). Read 2026-10-23.
- The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect of vitamin E and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J Med 1994. Read 2026-10-23.
- Omenn GS, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996 (CARET). Read 2026-10-23.
- Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20,536 high-risk individuals. Lancet 2002. Read 2026-10-23.
- Lonn E, et al. Effects of long-term vitamin E supplementation on cardiovascular events and cancer: a randomized controlled trial. JAMA 2005 (HOPE and HOPE-TOO). Read 2026-10-23.
- Klein EA, et al. Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA 2011. Read 2026-10-23.
Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.