Vision Supplement Guide: What The Evidence Shows

Almost every bottle sold for the eye traces its reputation to two studies, and almost none of them carries the amounts those studies used. Knowing what the trials were is therefore the single most useful thing a buyer in this category can know.

AREDS, published in 2001, randomised people with age-related macular degeneration to vitamin C 500 mg, vitamin E 400 IU, beta-carotene 15 mg, zinc 80 mg as zinc oxide and copper 2 mg as cupric oxide. It ran for more than six years and reported a reduction in progression to advanced disease in the higher-risk groups.

AREDS2, published in 2013, revised that formulation. It replaced the beta-carotene with lutein 10 mg and zeaxanthin 2 mg, tested DHA 350 mg with EPA 650 mg alongside, and tried zinc at 25 mg against 80 mg. The National Eye Institute publishes both formulations ingredient by ingredient and is the best free source for them.

Both columns are the trial formulations as published. Neither is a recommendation, and neither was tested in people with healthy eyes.

Two qualifications travel with those numbers and both are routinely dropped by the pages that quote them. The trials were run in people who already had macular disease, so the result is about slowing progression rather than about preventing anything. And the amounts are large: 500 mg of vitamin C is roughly eight oranges, and 80 mg of zinc is seven times a Daily Value.

The 2023 Cochrane review is the best summary of what those formulations achieved, and its companion on prevention is the harder read: in people who do not have the condition, the evidence that these nutrients prevent it is not there.

These two carotenoids are the reason AREDS2 exists and the reason most modern eye formulas name them on the front of the bottle. They are worth understanding properly, because what they do is specific rather than general.

Both concentrate in the macula, the small central part of the retina responsible for detailed vision, where they form what is called macular pigment. That pigment absorbs short-wavelength light and behaves as an antioxidant in the tissue that needs one most.

The measurable effect of supplementing them is a rise in macular pigment optical density, which a meta-analysis of the trials covers across doses and preparations. Whether that rise translates into a difference somebody notices is a separate question, and it is where the literature gets thinner.

The trial closest to the audience most of these products are sold to is a 2024 study in young adults with long-term digital device use, which measured macular pigment and visual function together. The LIMPIA trial ran the same idea over six months with a macular pigment endpoint. LIMPIA's abstract describes its intervention only as two daily capsules, so its per-capsule amounts are not quoted here; the 10 mg and 2 mg figures come from AREDS2 alone.

Zinc concentrates in the retina and the choroid, which is the tissue-level reason it appears in every formula in this category. AREDS gave it at 80 mg a day as zinc oxide and published a separate report on what five years at that amount did to serum zinc and to haematocrit.

AREDS2 then tested 25 mg against 80 mg and found the lower amount performed as well. That is the more useful number for reading an ordinary label, because 25 mg is an amount a supplement might plausibly carry and 80 mg is not.

Copper is on both formulations for one reason: sustained high-dose zinc depletes it. The mechanism is competitive absorption and the consequence is a copper deficiency that can present neurologically, which pharmacovigilance data tracks across large databases.

The antioxidant vitamins are the oldest part of this category and the part where the advice has moved most.

Vitamin C at the AREDS amount is 500 mg a day, which is more than five Daily Values and roughly eight oranges. Vitamin E was given as 400 IU of all-rac-alpha-tocopheryl acetate, the synthetic form, at about twenty-seven times its Daily Value.

Beta-carotene is the ingredient that was removed, and the reason is not ophthalmic. The ATBC trial gave 20 mg a day to 29,133 male smokers and reported more lung cancer in the beta-carotene arm. CARET gave 30 mg with retinyl palmitate and was stopped twenty-one months early for the same signal, which its own risk-factor analysis examines. A 2022 meta-analysis reaches the same conclusion across the trial literature.

Vitamin E's own high-dose literature runs in a similar direction. The SELECT trial gave 400 IU a day, the AREDS amount, to 35,533 men and reported a higher rate of prostate cancer in the vitamin E arm over a decade of follow-up.

None of that argues against the modest amounts found in an ordinary multivitamin. It argues against assuming that more of an antioxidant is better, which was the premise the whole category started from and which the trials have now tested twice.

A large part of this category carries plant extracts, usually inside a proprietary blend, and the honest description of that evidence is short.

Very few botanicals have a published trial measuring a vision outcome. What many of them have instead is a substantial literature measuring something else entirely, and the gap between those two things is where a lot of marketing lives.

None of the six has a published trial measuring a vision outcome. They are listed because they appear on eye formulas, not because their evidence applies to one.

That table gives a reader one rule worth carrying. When a botanical appears on an eye label, ask what its own trials measured. If the answer is prostate symptoms or hair growth, the ingredient may still be harmless and the argument for it is traditional rather than clinical.

The second rule is about weight. A blend declared at a couple of hundred milligrams across a dozen plants cannot carry a trial dose of any one of them, because several of those trial doses are larger than the blend.

Most people buying a vision supplement are not managing a diagnosed eye disease. They are uncomfortable at the end of a screen day, and that complaint has been studied on its own terms.

A 2026 review describes digital eye strain as two problems wearing one name. The surface problem is a drop in blink rate and a tear film that breaks up faster. The focusing problem is a ciliary muscle holding one near distance for hours at a time.

Neither is nutritional, which is the uncomfortable sentence in a guide on a supplement website. The measures with direct evidence behind them are behavioural and optical.

• Breaks from near work. The 20-20-20 trial tested the best-known version in symptomatic computer users over two weeks against their own baseline, and a companion paper asks fairly where the three numbers came from.
• An up-to-date prescription. An uncorrected small refractive error is the commonest reason a screen day is exhausting, and it is invisible to the person having it.
• Screen position below eye level, which reduces the exposed area of the eye surface and so the rate at which the tear film evaporates.
• Deliberate blinking and room humidity, both of which act on the surface half of the problem.
• Lubricating drops where the surface is genuinely dry, which the American Academy of Ophthalmology covers alongside the rest.

A supplement can sit alongside all five and none of them costs anything. A supplement taken instead of them is the expensive half of the answer on its own, and any honest guide in this category has to say so.