Can Red Light Therapy Improve Eyesight & Is It Safe for the Eyes?

Two questions come up together: can red light do anything for eyesight, and is it safe to use a panel with your eyes open? The honest answer to the first is "it has been studied in specific eye conditions, with specific devices, and some of those trials are encouraging". The honest answer to the second is longer, so it gets its own section.

What has actually been studied

The strongest human evidence is in dry age-related macular degeneration. LIGHTSITE III, a randomised sham-controlled trial of 100 people, delivered 590, 660 and 850nm light in nine-session courses every four months and found a 5.4-letter gain in visual acuity at 13 months against 3.0 letters for sham, with fewer treated eyes developing new geographic atrophy and a favourable safety profile. The gap of 2.4 letters is real but modest, and five of the authors work for the device company. A second device, using 630nm red with 590nm yellow, was tested in 76 patients: two sessions a week for four weeks, with a significant change in best-corrected acuity at four months and only mild ocular adverse events in a fifth of the treated group. Earlier and smaller: 670nm LED light reduced drusen volume and improved acuity and contrast sensitivity in 42 eyes, and nine 670nm sessions improved acuity and visual field in a short-term study.

A 2023 narrative review of all of this is worth reading for its caution: early results for acuity and contrast sensitivity are encouraging, but with only three randomised trials, varied protocols and small samples, whether the treatment slows progression to advanced AMD is still uncertain. Beyond AMD, twelve people with retinitis pigmentosa had nine sessions of the same three wavelengths: acuity improved from 0.62 to 0.53 logMAR four weeks later, visual field measures were numerically but not significantly better, and there were no adverse effects on OCT or ERG. An open study, so no sham arm. Our eyesight research page lists all of these by wavelength.

The myopia work is a different device

You may have seen headlines about red light slowing short-sightedness in children. That work, mostly from China, uses a different thing: a 2022 review describes a 650nm low-level laser at about 2 mW that the child looks into for a few minutes a day, and concludes it may be a useful tool while calling for more studies. It is a laser aimed into the eye, not a panel across a room, and it has a published cautionary case: retinal damage in a 12-year-old after repeated low-level red-light laser exposure. We mention it because the two get mixed up, and because it is the clearest reason not to point any laser at your eyes without an ophthalmologist involved.

What the lab studies add

Cell and animal work explains why researchers went looking. 670nm light increased the ability of human retinal pigment cells to clear debris; ten minutes of 810nm or 670nm light protected mouse retinal tissue after six hours of damaging blue light, boosting mitochondrial energy metabolism and cutting cell death; and daily 830nm LED sessions preserved photoreceptors in a rat model of retinitis pigmentosa, with the benefit fading once treatment stopped. These are mechanism, not benefit in people. They are also the reason the eye trials chose the wavelengths they did.

A curious one: light on the back

In 2025 a group of vision researchers put 25 adults under an 850nm LED panel for 15 minutes with the light on the bare back, and measured colour vision a day later: blue-yellow colour-contrast thresholds improved by 16%, and still by 7% in a group whose heads were wrapped in foil so no light reached the eyes. Sham exposure did nothing. The authors read it as a body-wide mitochondrial effect; it is one study of 25 people and needs repeating, but it is the first hint that a panel session might reach the eyes without the eyes being in the light at all.

Is it safe for the eyes?

What we can say with sources: the eye trials above aimed light into the eye at set doses and reported good tolerability, from a favourable safety profile over 13 months to no adverse effects on OCT or ERG. Red and near-infrared also sit at the bottom of the retinal hazard scale that lighting standards use - the one that puts blue near the top - which we explained in the blue light post.

What we cannot say: nobody has followed panel users' eyes over years, because nobody has done that study. A panel is far brighter to look at than a clinic device, and the natural reaction is to close your eyes. So every panel ships with goggles, spare eye protection is on the site, and the decision belongs with you and your eye specialist - particularly if you have an eye condition, take medication that increases light sensitivity, or are using a laser device of any kind. An earlier version of this post cited a review of light therapy for mood disorders as evidence of eye safety; it was about bright white light boxes, not red panels, so it is gone.

What we suggest

Wear the goggles or close your eyes in front of a panel, keep lasers away from your eyes unless an ophthalmologist is involved, and if you have an eye condition, ask your specialist before you start. If you want the studies, the eyesight research page has them; if you want to ask, contact us.