Oxidative stress and antioxidants: why more is not better

Oxidative stress is real, but the popular story built around it is mostly wrong. The idea that ageing and disease come from free radicals, and that swallowing extra antioxidants will mop them up and keep you young, has been tested in very large randomised trials and has failed repeatedly. In some, high-dose antioxidant supplements did not just fail to help: they raised the risk of cancer or death. The honest position is that reactive oxygen species are essential signalling molecules, your own defences are far more sophisticated than any pill, and the antioxidants worth caring about come from food, not megadose capsules.

What oxidative stress actually is

Reactive oxygen species (ROS) are small, chemically reactive molecules made mostly as a by-product of using oxygen to generate energy in your mitochondria. The best-known examples are the superoxide radical, hydrogen peroxide and the hydroxyl radical. A free radical is simply a molecule with an unpaired electron, which makes it eager to react with whatever is nearby, including DNA, proteins and the fats in your cell membranes. Oxidative stress is the name for an imbalance: more reactive species being produced than your defences can manage, leaving a residue of molecular damage.

So far this sounds like a clean villain story, and for decades it was told that way. The twist is that ROS are not only damage agents. At low, controlled levels they are signalling molecules: the cell uses bursts of hydrogen peroxide and superoxide to regulate metabolism, immune responses, blood-vessel tone and how cells adapt to stress. This dual role, useful messenger at low doses and harmful at high ones, is the fact the supplement industry tends to skip. Flood the system with antioxidants and you are not just removing damage; you may be silencing signals the body needs. For how cellular wear-and-tear maps onto ageing, see our guide to the hallmarks of ageing.

Why the free-radical theory of ageing fell apart

The free-radical theory of ageing was proposed by Denham Harman in 1956. It held that ageing is the cumulative result of free-radical damage, and the obvious implication was that antioxidants should slow ageing. It is a beautiful, simple idea, and it drove an entire supplement category. The problem is that when researchers tried to confirm it, the data kept refusing to cooperate.

Several findings broke the simple version. Calorie restriction and regular exercise both extend healthy lifespan in animals, yet both can increase ROS production rather than reduce it. Genetically boosting antioxidant enzymes in mice often failed to extend lifespan. And the large human trials of antioxidant pills, covered below, showed no benefit and sometimes harm. Reviewers now describe the original theory as largely superseded, replaced by a model built around redox signalling and hormesis.⁷

Hormesis is the principle that a small dose of a stressor triggers an adaptive response that leaves the organism more resilient than before. A brief, mild rise in mitochondrial ROS during exercise, sometimes called mitohormesis, is now understood to be one of the signals driving the benefits of training. A controlled bit of oxidative stress is part of how exercise makes you healthier; blanket-suppressing it with antioxidants removes the trigger. This is why "more is better" is exactly the wrong instinct.

The big supplement trials, and how badly they went

This is the part that should change how you shop. Because people who eat more fruit and vegetables get less cancer and heart disease, and those foods are rich in antioxidants, researchers ran enormous randomised trials giving the isolated antioxidants as supplements. If the free-radical theory were right, these should have been triumphs. They were the opposite.

The ATBC trial (Finland, published 1994 in the New England Journal of Medicine) randomised 29,133 male smokers to alpha-tocopherol (vitamin E, 50 mg/day), beta-carotene (20 mg/day), both or placebo. Beta-carotene did not prevent lung cancer; if anything, lung cancer incidence was higher in the beta-carotene group.¹ The American CARET trial (1996, also NEJM) tested beta-carotene plus vitamin A in 18,314 smokers and asbestos-exposed workers. It was stopped 21 months early because the supplement group had 28% more lung cancers and 17% more deaths.² Two separate trials, the same uncomfortable direction.

Vitamin E fared no better. The HOPE and HOPE-TOO trials gave 400 IU/day of natural vitamin E to people at high cardiovascular risk for a median of 7 years. There was no protection against cancer or major cardiovascular events, and a higher rate of heart failure in the vitamin E group.⁹ Then came SELECT (published in JAMA in 2011), which randomised 35,533 healthy men to vitamin E, selenium, both or placebo. Vitamin E significantly increased prostate cancer risk by roughly 17%: about 76 cases per 1000 men on vitamin E versus 65 per 1000 on placebo.³

Pulling it together, the 2012 Cochrane review by Bjelakovic and colleagues analysed 78 randomised trials covering 296,707 participants. In the trials with the lowest risk of bias, beta-carotene and vitamin E significantly increased all-cause mortality, and higher doses of vitamin A tracked with higher mortality.⁴ For these isolated, high-dose antioxidants, the supplement story did not merely fail to deliver, it backfired.

The exercise problem: antioxidants can blunt your gains

If you train, this section matters more than the cancer trials. Because exercise generates a transient rise in ROS, and because that rise is part of the adaptation signal, dosing yourself with antioxidants around training can interfere with the very benefits you are exercising for.

In a 2009 study published in PNAS, Michael Ristow and colleagues gave young men vitamin C (1000 mg/day) plus vitamin E (400 IU/day) or placebo during a four-week exercise programme. Exercise improved insulin sensitivity only in the men not taking the antioxidants; the supplements blocked the metabolic benefit and suppressed the cell's own antioxidant-enzyme response.⁵ A 2014 randomised trial in The Journal of Physiology by Gøran Paulsen and colleagues gave 54 adults vitamin C (1000 mg) and vitamin E (235 mg) or placebo during 11 weeks of endurance training. The supplemented group showed blunted increases in markers of mitochondrial biogenesis such as PGC-1-alpha and COX4, the cellular machinery that makes you fitter.⁶

The practical message is not "never take vitamin C". It is that chronic high-dose antioxidant supplements taken to support training may quietly work against it. If you exercise to improve metabolic health or endurance, the evidence favours getting antioxidants from food and letting the training signal run its course. Our insights pieces on training adaptation go further into this.

Your built-in defences: the Nrf2 system

One reason swallowing antioxidants is a blunt instrument is that your body already runs a sophisticated, self-regulating defence. At its centre is a transcription factor called Nrf2, held in check by a partner protein, Keap1. Keap1 acts as a chemical sensor: when it detects oxidative or electrophilic stress, it releases Nrf2, which moves into the nucleus and switches on a whole battery of protective genes.⁸

Those genes produce your endogenous antioxidants and detox enzymes: glutathione-making machinery, superoxide dismutase, catalase, glutathione peroxidase and more. This is far cleaner than a steady external dose, because it is responsive: defences ramp up where they are needed and settle when they are not. Crucially, Nrf2 is activated by mild stressors, including exercise and certain plant compounds such as the sulforaphane in broccoli sprouts. So some of the genuine benefit attributed to "antioxidant" foods may come not from molecules directly neutralising radicals, but from them gently nudging your own defence system into action, a hormetic effect again. A constant high dose of a simple antioxidant does not engage this adaptive machinery in the same way.

Why food antioxidants are not the same as pills

It is tempting to assume that if vegetables help and vegetables contain antioxidants, then the antioxidants in a capsule should help too. The trials above show that logic fails. Here is why.

• Dose and form differ enormously. A supplement may deliver many times the amount of a single compound found in any plate of food, in an isolated form the body rarely meets in nature. The trials that harmed people used pharmacological doses, not dietary ones.
• Whole foods are a package, not a molecule. Fruit, vegetables, herbs, tea and coffee deliver hundreds of polyphenols and other compounds together with fibre and minerals, which interact in ways an isolated vitamin cannot reproduce.
• Many food antioxidants are poorly absorbed, and that may be the point. Polyphenols have famously low bioavailability: most never reach high blood concentrations.¹⁰ Rather than acting as direct radical scavengers in the bloodstream, much of their benefit appears to come from low-level signalling, including Nrf2 activation, and from their effects in the gut.
• Confounding cuts both ways. People who eat more plants also tend to smoke less, move more and be wealthier, so some of the observed benefit of "antioxidant-rich diets" is not antioxidants at all. This is precisely why isolating the supposed active ingredient into a pill was such an appealing test, and why its failure is so informative.

The takeaway is genuinely food-first: a varied diet rich in vegetables, fruit, legumes, nuts and herbs gives you these compounds in the context the evidence actually supports. If you are deciding which supplements, if any, are worth your money, our stack builder helps you avoid paying for things trials have shown do not work.

The honest bottom line

Oxidative stress is a real process, but it is not a simple poison to be neutralised. Reactive oxygen species are essential signals, your Nrf2-based defences are smarter than any pill, and a little oxidative stress from exercise is part of how you get healthier. The free-radical theory of ageing has been overtaken by a redox-signalling and hormesis model that explains why "mop up the radicals" was always too crude. Most importantly, the largest and best human trials of high-dose antioxidant supplements have shown no benefit and, for beta-carotene and vitamin E, real harm. Eat the rainbow, train, sleep, and be sceptical of any product promising to fight ageing by flooding you with antioxidants. New to this? Our getting-started guide covers how to evaluate a supplement claim before you spend a penny.

This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.