The AREDS2 Formula, Dose by Dose — and Where Everyday Eye Supplements Diverge From It

The AREDS2 formula, dose by dose

Two National Eye Institute trials produced the numbers everybody quotes. The first, published as AREDS Report No. 8 in Archives of Ophthalmology in 2001, tested a combination of antioxidant vitamins plus zinc. The second, AREDS2, published in JAMA in 2013, reworked the ingredient list. Below is what was actually inside each capsule, in the amounts the investigators used.

Daily amounts as tested. Sources: AREDS Report No. 8, Arch Ophthalmol 2001;119(10):1417-36 (PMID 11594942); AREDS2 Research Group, JAMA 2013;309(19):2005-15 (PMID 23644932).

Nutrient	AREDS (2001)	AREDS2 (2013)	AREDS2 low-zinc arm
Vitamin C	500 mg	500 mg	500 mg
Vitamin E	400 IU	400 IU	400 IU
Beta-carotene	15 mg	Removed	Removed
Lutein	Not included	10 mg	10 mg
Zeaxanthin	Not included	2 mg	2 mg
Zinc (as zinc oxide)	80 mg	80 mg	25 mg
Copper (as cupric oxide)	2 mg	2 mg	2 mg

Read the table across rather than down and the design becomes obvious. Only two rows ever move. Vitamin C, vitamin E and copper are constant across every version ever tested. Beta-carotene is the row that disappears. Lutein and zeaxanthin are the rows that appear. Zinc is the row that has two numbers because AREDS2 deliberately tested both.

One labelling note that trips people up: the 400 IU of vitamin E in both trials was synthetic dl-alpha-tocopheryl acetate, which is roughly 268 mg of alpha-tocopherol. Modern labels are required to print milligrams, so a product matching the trial amount will show something close to 268 mg rather than the familiar 400 IU. The adult RDA for vitamin E is 15 mg; the tolerable upper intake level is 1,000 mg. The trial dose sits well above the first and well below the second.

What each trial arm was designed to answer — and what it found

The formula table tells you what was in the capsule. It does not tell you what question each capsule was built to answer, which is where most write-ups stop and where the useful detail lives. AREDS2 was not one experiment. It was a primary randomisation of 4,203 people into four groups, plus a secondary randomisation of the 3,036 participants (72 per cent) who consented to it.

Trial arms, the question each was built to answer, and the published result.

Trial / arm	Question it was designed to answer	What it found
AREDS 2001 — antioxidants + zinc vs placebo (n=3,640, mean 6.3 years)	Does a high-dose antioxidant and zinc combination change progression to advanced AMD?	Odds ratio 0.72 for progression to advanced AMD (99% CI 0.52–0.98), and 0.73 for losing 15 or more letters of visual acuity (99% CI 0.54–0.99).
AREDS2 primary — lutein 10 mg + zeaxanthin 2 mg vs placebo (n=4,203, median 5 years)	Does adding the two macular carotenoids to the AREDS base improve on it?	No statistically significant improvement over placebo on the primary endpoint. Hazard ratio 0.90 (98.7% CI 0.76–1.07, P=.12). Five-year progression was 29% with lutein and zeaxanthin versus 31% with placebo.
AREDS2 primary — DHA 350 mg + EPA 650 mg vs placebo	Do marine omega-3 fatty acids add anything?	No. Hazard ratio 0.97 (98.7% CI 0.82–1.16, P=.70). Omega-3s were not carried into the final formula.
AREDS2 secondary — zinc 25 mg vs 80 mg (n=3,036)	Is the original 80 mg zinc dose necessary?	No significant difference in AMD outcomes between the two doses.
AREDS2 secondary — with vs without beta-carotene	Can beta-carotene be removed without losing the effect?	No apparent loss of effect. More lung cancers were recorded in the beta-carotene group (23, or 2.0%, versus 11, or 0.9%; nominal P=.04), mostly in former smokers.
AREDS2 Report 4 — cataract surgery endpoint (n=3,159 phakic in at least one eye, median 4.7 years)	Does lutein and zeaxanthin change rates of cataract surgery?	No significant overall effect, hazard ratio 0.96 (95% CI 0.84–1.10, P=.54). In the subgroup with the lowest dietary carotenoid intake, hazard ratio 0.68 (95% CI 0.48–0.96, P=.03).
AREDS2 Report 28 — 10-year follow-on (n=3,882, mean age 72.0)	Over a decade, how does lutein and zeaxanthin compare with beta-carotene?	Hazard ratio 0.85 for progression to late AMD versus beta-carotene (95% CI 0.73–0.98, P=.02). Lung-cancer odds ratio 1.82 for beta-carotene (95% CI 1.06–3.12, P=.02) versus 1.15 for lutein and zeaxanthin (95% CI 0.79–1.66, P=.46).

That second row is the one nobody puts on a label. In the headline AREDS2 comparison, adding lutein and zeaxanthin to the AREDS base did not beat placebo at the pre-specified level of significance. The case for the swap is built on the secondary and long-term analyses instead: the ten-year Report 28 found lutein and zeaxanthin came out ahead of beta-carotene on late AMD progression, and beta-carotene carried a lung-cancer signal that the carotenoids did not. That is a safety-and-substitution argument, not a potency argument, and it is worth knowing which one you are being sold. We look at the efficacy question itself in more depth in our review of what the lutein and zeaxanthin evidence actually shows.

Why 80 mg of zinc, and why there is copper in the capsule

The 80 mg zinc figure has an awkward property: it is double the tolerable upper intake level. The Institute of Medicine sets the adult UL for zinc at 40 mg per day, against an RDA of 11 mg for men and 8 mg for women. AREDS chose 80 mg in the 1990s on the strength of earlier work suggesting zinc mattered to retinal enzyme function, and AREDS2 inherited it.

Two facts belong in the same sentence and almost never are. First, 80 mg is twice the published upper limit for adults. Second, when AREDS2 finally tested 25 mg against 80 mg head to head in 3,036 people, it found no significant difference in macular outcomes. Both are true. They are the reason a large share of products sold as "AREDS2 formulas" today carry 25 mg or 40 mg of zinc rather than 80 mg, and the reason the two numbers you see on shelves are not a sign that one manufacturer is cutting corners. Which of the two figures is the better buy is argued out in how much zinc an eye supplement actually needs, and the tolerability cost of carrying a high-zinc formula every day is covered in eye supplement side effects.

Who the AREDS trials actually enrolled

This is the paragraph that decides whether the rest of the article is relevant to you. AREDS enrolled 3,640 participants aged 55 to 80 and followed them for an average of 6.3 years. AREDS2 enrolled 4,203 participants aged 50 to 85 and followed them for a median of five years, with a subset carried to ten years in Report 28. In both trials, entry required existing macular disease: intermediate age-related macular degeneration in both eyes, or advanced disease in one eye and intermediate in the other.

Nobody in either trial had healthy eyes. That is not a technicality. It means the published effect sizes describe what happened to progression risk in a population that already had measurable macular change, over five to ten years. It does not describe what happens to a 45-year-old with a clean retinal exam who is tired of screen glare, because that person was never in the study. Neither trial measured whether the formula sharpens ordinary eyesight, and neither one reported that it did.

It also means the results are structurally hard to extrapolate. Applying a hazard ratio measured in people with intermediate AMD to people without it is not a conservative assumption; it is an unsupported one in either direction. The honest position is that the AREDS trials answer a specific clinical question for a specific clinical population, and that anyone who thinks they belong to that population should be discussing it with an ophthalmologist rather than with a website.

Where an everyday carotenoid capsule diverges from a clinical AREDS2 product

There are two different product categories on the shelf and they are frequently confused.

A clinical AREDS2 product exists to reproduce the trial formulation as closely as possible for people who have been told by an eye doctor that they have intermediate AMD. Its whole design brief is fidelity to the published amounts, which is why those labels lead with the numbers and why they carry high-dose zinc with the matching copper.

An everyday carotenoid capsule — the category Visivra sits in — is built around a different intent. It typically supplies lutein and zeaxanthin alongside supporting antioxidants at general-nutrition amounts, and it is aimed at people who simply eat few dark leafy greens. It is not a trial replica and should not be read as one. The practical differences to look for on a panel are: whether lutein and zeaxanthin are declared in milligrams rather than buried in a proprietary blend, whether zinc is anywhere near a trial amount or at ordinary multivitamin levels, whether copper is present if the zinc is high, and whether the antioxidant vitamins are at the 500 mg and 268 mg trial amounts or at typical RDA-level doses.

We deliberately do not publish milligram figures for Visivra on this site, because we have not been able to verify a Supplement Facts panel from the manufacturer's own page. Where we do not have a verified dose, we say so. Read the label on the bottle and check it against the trial column yourself — that is the entire point of having the table above.

What the AREDS evidence does not show

Four honest limits, stated plainly.

It does not show that the formula helps people without macular disease. No such population was enrolled. There is no arm of either trial that answers the question.

It does not show that lutein and zeaxanthin beat placebo on the AREDS2 primary endpoint. They did not, at the pre-specified significance level. The substitution case rests on the ten-year comparison against beta-carotene and on the safety signal, both of which are real, and neither of which is the same claim.

It does not show that any commercial product performs like the trial formula. A finished supplement is not a clinical intervention. Ingredients can have evidence while a specific product has never been trialled at all, and that is the ordinary state of affairs in this category.

It does not show that the trial amounts are safe for everyone. Zinc at 80 mg exceeds the adult upper limit. Beta-carotene at 15 mg carried a lung-cancer signal in smokers and former smokers, which is precisely why it was removed. Trial doses were given under supervision with monitoring; a bottle bought online comes with neither.

Where food fits into all of this is a separate question with its own numbers — we work through it in how much spinach actually equals a 10 mg lutein capsule.

Frequently asked questions