AREDS supplements slow advanced macular degeneration. They do not prevent AMD.

The AREDS formulation reduces the odds of progressing to advanced age-related macular degeneration in people who already have intermediate AMD, or advanced AMD in one eye — by about a quarter, in the original trial [s1]. It has never been shown to prevent AMD in anyone who does not already have it, because that population was not the one enrolled. And the follow-up trial that tested the two ingredients most heavily marketed since — lutein with zeaxanthin, and omega-3 fatty acids — found no statistically significant benefit from either on its primary endpoint [s2].

What the original trial actually tested

The Age-Related Eye Disease Study was an 11-centre double-masked randomised trial of 3,640 participants aged 55 to 80, followed for an average of 6.3 years with 2.4% lost to follow-up [s1]. Participants were assigned to one of four daily regimens: antioxidants (vitamin C 500 mg, vitamin E 400 IU, beta carotene 15 mg); zinc 80 mg as zinc oxide with copper 2 mg as cupric oxide; antioxidants plus zinc; or placebo [s1].

Against placebo, antioxidants plus zinc produced a statistically significant reduction in the odds of developing advanced AMD: odds ratio 0.72 (99% confidence interval 0.52 to 0.98) [s1]. Zinc alone (OR 0.75, 99% CI 0.55 to 1.03) and antioxidants alone (OR 0.80, 99% CI 0.59 to 1.09) did not reach significance [s1]. On the second primary outcome — losing 15 letters or more of visual acuity — only the combined antioxidants-plus-zinc arm showed a significant reduction, OR 0.73 (99% CI 0.54 to 0.99) [s1].

The most useful number in the paper is the one least often quoted. Among the 1,063 participants whose eyes showed only extensive small drusen, non-extensive intermediate drusen or pigment abnormalities, the five-year probability of progressing to advanced AMD was 1.3% [s1]. There was almost nothing for a supplement to prevent. When those participants were excluded, the effect estimates in the remaining higher-risk group strengthened — antioxidants plus zinc to OR 0.66 (99% CI 0.47 to 0.91), zinc alone to 0.71 (99% CI 0.52 to 0.99) [s1].

That is the whole basis for the standard caveat that AREDS supplements are for people with intermediate or advanced disease. It is not a conservative reading of the trial; it is what the trial found.

AREDS2 and the two null results

AREDS2 enrolled 4,203 participants aged 50 to 85 between 2006 and 2012, all at risk of progression with bilateral large drusen, or large drusen in one eye and advanced AMD in the other [s2]. They were randomised to lutein 10 mg plus zeaxanthin 2 mg, DHA 350 mg plus EPA 650 mg, both, or placebo, on top of the original AREDS formulation or a variation of it [s2].

Over a median five years, 1,940 study eyes in 1,608 participants progressed to advanced AMD [s2]. The five-year Kaplan-Meier probabilities were 31% on placebo, 29% on lutein plus zeaxanthin, 31% on DHA plus EPA, and 30% on both [s2]. None of the primary comparisons reached significance: hazard ratio 0.90 (98.7% CI 0.76 to 1.07; P = .12) for lutein plus zeaxanthin, 0.97 (98.7% CI 0.82 to 1.16; P = .70) for DHA plus EPA, and 0.89 (98.7% CI 0.75 to 1.06; P = .10) for both [s2]. Eliminating beta carotene or lowering the zinc dose had no apparent effect on progression either [s2].

The reason lutein and zeaxanthin nonetheless ended up in the modern formulation was a safety finding, not an efficacy one. More lung cancers occurred in the beta carotene group than the no-beta-carotene group — 23 (2.0%) versus 11 (0.9%), nominal P = .04 — mostly among former smokers [s2]. Lutein and zeaxanthin were an appropriate substitute for a carotenoid that was doing harm, which is a different claim from being an effective addition.

What ten years of follow-up added

An epidemiological follow-up of the AREDS2 cohort, conducted from December 2012 to December 2018 and covering 3,882 participants (mean baseline age 72.0 years, SD 7.7; 2,240 women, 57.7%) and 6,351 eyes, reported ten-year outcomes [s3]. The lung cancer signal held: the odds ratio at ten years was 1.82 (95% CI 1.06 to 3.12; P = .02) for those randomised to beta carotene, against 1.15 (95% CI 0.79 to 1.66; P = .46) for lutein/zeaxanthin [s3].

On AMD progression, the longer horizon produced a marginal result for lutein/zeaxanthin that the five-year trial had not: hazard ratio 0.91 (95% CI 0.84 to 0.99; P = .02) [s3]. Omega-3 fatty acids remained flatly null (HR 1.01, 95% CI 0.93 to 1.09; P = .91), as did low versus high zinc (HR 1.04, 95% CI 0.94 to 1.14; P = .49) [s3]. The strongest-looking lutein result came from a restricted comparison — among those also randomised to beta carotene, HR 0.80 (95% CI 0.68 to 0.92; P = .002) — and in a direct head-to-head against beta carotene the HR for late AMD was 0.85 (95% CI 0.73 to 0.98; P = .02) [s3]. Those are comparisons against an ingredient now removed, in an observational extension of a trial, and the authors framed the conclusion as lutein/zeaxanthin being an appropriate replacement for beta carotene rather than as an established benefit [s3].

What none of this shows

Three claims commonly attached to eye vitamins are not supported by these trials. They were not tested in people with healthy retinas, so nothing here speaks to prevention in the general population. They were not shown to restore vision — the endpoints were progression and acuity loss, not improvement. And the doses are far above dietary intake: 500 mg of vitamin C, 400 IU of vitamin E and 80 mg of zinc are not the amounts in a general multivitamin, and the zinc dose in particular was chosen for the trial rather than from a nutritional requirement [s1].

What the evidence does support is narrow and durable: in people whose retinas already show the specific changes AREDS enrolled, a high-dose formulation reduced the odds of progressing to advanced disease over roughly six years [s1], and the reformulation that replaced beta carotene with lutein and zeaxanthin removed a demonstrated lung cancer risk without losing that benefit [s2] [s3].

Whether any individual has the retinal changes those trials selected for is determined by a dilated eye examination, not by symptoms. This article describes trial results and is not medical advice; supplement doses at these levels interact with other conditions and medications and are a matter for a clinician.