Blue Light Glasses: 17 Trials, and None Measured Your Retina

There is a particular shape to this evidence that is worth seeing, because it is not the usual “studies are mixed” story.

Cochrane reviewers went looking for randomised trials of blue-light filtering spectacle lenses against non-filtering lenses, with no date or language restriction, searching CENTRAL, MEDLINE, Embase, LILACS, the ISRCTN registry, ClinicalTrials.gov and the WHO registry. They last searched in March 2022.

They found 17 trials. Then they declined to pool them.

The meta-analysis that did not happen

This is the first finding, and most coverage skips it.

A Cochrane review normally combines trial results into a single pooled estimate. Here the reviewers wrote that they “did not perform meta-analyses for any of the outcome measures, due to lack of available quantitative data, heterogenous study populations, and differences in intervention follow-up periods.”

Not one outcome could be pooled. Seventeen trials, and the data would not carry a combined number for any of them.

The trials themselves explain why. Sample sizes ran from five participants to 156. Follow-up ran from less than a single day to five weeks. Populations ranged from healthy adults to people with mental health and sleep disorders. About half were parallel-arm, the rest cross-over.

What the trials were worth

Cochrane assessed each trial against seven risk-of-bias domains. The result:

• None of the 17 studies had a low risk of bias across all seven domains
• 65% were judged at high risk of detection bias, because outcome assessors were not masked
• 59% were at high risk of performance bias, because participants and personnel were not masked
• 35% were pre-registered on a trial registry

Masking matters more here than in most fields. If you can see that your lenses have a tint, and you know the tint is the thing being tested, you are no longer a blinded participant. Roughly two-thirds of this literature has that problem.

Eye strain: the confidence interval tells the story

The review’s primary outcome was change in visual fatigue score. One trial of 120 participants reported a mean difference of 9.76 units in the direction of worse symptoms — with a 95% confidence interval running from -33.95 to 53.47.

Read that interval rather than the point estimate. It spans a large benefit and a large harm. It is consistent with the glasses helping considerably, hurting considerably, and doing nothing at all.

Two further studies, 46 participants combined, reported no significant difference. Cochrane rated the evidence low certainty and concluded there “may be no difference in subjective visual fatigue scores.”

The authors’ summary sentence: blue-light filtering lenses “may not attenuate symptoms of eye strain with computer use, over a short-term follow-up period.”

Vision: exactly zero

One outcome came out clean, and it is the one nobody argues about.

For best-corrected visual acuity, one study of 156 participants found a mean difference of 0.00 logMAR, 95% CI -0.02 to 0.02. Cochrane rated this moderate certainty — the highest certainty rating anywhere in the review — and concluded there is “probably little or no effect.”

So the lenses do not blur your vision. That is worth knowing if you already own a pair. It is not what they are sold for.

Sleep: three said yes, three said no

Six randomised trials, 148 participants between them, looked at sleep quality.

Three reported a significant improvement with blue-light filtering lenses. Three reported no significant difference.

Cochrane rated this very low certainty and wrote: “We do not know if blue-light filtering spectacle lenses are equivalent or superior to non-blue-light filtering spectacle lenses with respect to sleep quality.”

An even split across 148 people, in trials mostly not masked, is not a mixed result that further reading will resolve for you. It is an open question with very little in it.

The retina claim has no trials at all

This is the finding that inverts the marketing.

Blue-light lenses are frequently sold as protecting the macula from screen light. Cochrane’s sentence on this is flat:

“There was no evidence from RCT publications relating to the outcomes of contrast sensitivity, colour discrimination, discomfort glare, macular health, serum melatonin levels, or overall patient visual satisfaction.”

Six outcomes. Zero trials each.

Note which two are on that list. Macular health — the retinal protection claim. And serum melatonin — the biological mechanism the sleep claim rests on.

The two things the advertising is most confident about are the two the trial literature has never measured. Not measured and found wanting. Simply never measured.

The side effects nobody advertises

Nine trials, 333 participants, reported on adverse events; three of them described events occurring. Cochrane records that reported adverse events related to blue-light filtering lenses were infrequent, but included increased depressive symptoms, headache, discomfort wearing the glasses, and lower mood.

Adverse events with the ordinary lenses were occasional hyperthymia and discomfort wearing the spectacles.

Low-certainty evidence, inconsistently reported, and small numbers. Worth stating anyway, because a product marketed as having no downside is being sold to you as having no downside.

What this does not mean

It does not mean the glasses are proven useless. A review that could not pool a single outcome is telling you the evidence is too thin and too varied to support a conclusion in either direction. Cochrane closes by calling for “future high-quality randomised trials,” which is what you write when the question is open rather than settled.

It does not mean your eye strain is imaginary. It means this particular product has not been shown to fix it, and that the review compared only filtering lenses against non-filtering lenses. It tested nothing else, so it says nothing about screen breaks, blink rate, room lighting, or an uncorrected prescription. If your eyes are dry rather than strained, that is a different problem with its own evidence, and we went through the omega-3 dry eye trial where both groups improved separately.

And if you already own a pair, nothing here says stop. The one moderate-certainty finding in the whole review is that they do not degrade your corrected vision.

What the eye doctors say the problem actually is

Cochrane assessed a product. The American Academy of Ophthalmology addresses the underlying question, and its position is blunter than the review’s:

“Digital eye discomfort is not caused by blue light.”

The Academy states that “the small amount of blue light coming from computer screens has never been shown to harm your eyes,” that “blue light from computers will not lead to eye disease,” and that “so far, the evidence shows no meaningful link between blue light and damage to human retinas or age-related macular degeneration.”

It attributes the symptoms to “device misuse or overuse” instead, and names a mechanism you can check on yourself:

“We blink less when we stare at our devices. Normally, we blink around 15 times per minute – but this ‘blink rate’ can be cut in half when staring at screens.”

Fifteen blinks a minute, halved. That is a drying problem, not a wavelength problem, which is why a lens that filters wavelengths would not be expected to fix it.

On sleep the Academy is equally direct: “You don’t need to spend extra money on blue light glasses to improve sleep – simply decrease evening screen time and set devices to night mode.”

Why there is no product link on this page

Because the conclusion does not support a purchase.

We place product links on pages where something specific follows from what the evidence showed. Here what follows is that the strongest claim has no trials behind it, and pointing you at a checkout for a commission after writing that would make the rest of the page worthless.

The short version

Seventeen randomised trials. Zero meta-analyses, because the data would not support one. Zero of the 17 at low risk of bias across all seven domains. On eye strain, a confidence interval from -33.95 to 53.47. On vision, 0.00 logMAR. On sleep, three trials for and three against across 148 people. On your retina and on melatonin, no trials whatsoever.

That is not a product with weak evidence. On its headline claim, it is a product with none.

Frequently Asked Questions

Do blue light glasses actually reduce eye strain?

Cochrane’s 2023 review of 17 randomised trials concluded that blue-light filtering lenses “may not attenuate symptoms of eye strain with computer use, over a short-term follow-up period,” and rated that evidence low certainty. The single largest relevant trial, with 120 participants, reported a mean difference of 9.76 units in the direction of worse symptoms with a 95% confidence interval from -33.95 to 53.47 — an interval so wide it is compatible with meaningful benefit, meaningful harm and no effect at all. Two further studies covering 46 participants found no significant difference. The honest reading is that the question has not been answered, not that it has been answered no.

Do blue light glasses protect your retina?

There is no randomised trial evidence either way, which is a stronger statement than it sounds. Cochrane searched with no date or language restriction and reported that among the 17 included trials there was “no evidence from RCT publications relating to the outcomes of contrast sensitivity, colour discrimination, discomfort glare, macular health, serum melatonin levels, or overall patient visual satisfaction.” Macular health is the retinal protection claim, and not one of the 17 trials measured it. That is the gap between what these lenses are advertised to do and what has ever been tested.

Do blue light glasses help you sleep?

Cochrane rated the evidence very low certainty and wrote that “we do not know” whether the lenses are equivalent or superior to ordinary lenses for sleep quality. Six trials totalling 148 participants reported on it: three found a significant improvement, three found no significant difference. Notably, none of the 17 trials measured serum melatonin, which is the mechanism usually offered for why the glasses should work. So the proposed biological pathway is untested and the outcome it is supposed to produce is evenly split in small, mostly unmasked trials.

Why did Cochrane not combine the 17 trials into one number?

Because the trials were too different from one another and too short of usable numbers. The review states it “did not perform meta-analyses for any of the outcome measures, due to lack of available quantitative data, heterogenous study populations, and differences in intervention follow-up periods.” Sample sizes ranged from five participants to 156, follow-up ran from under one day to five weeks, and populations ranged from healthy adults to people with mental health and sleep disorders. When a Cochrane team declines to pool anything at all, that itself is the headline finding about the state of the literature.

Are blue light glasses bad for you?

Nothing in the review suggests they damage your eyes, and the one moderate-certainty finding is that they have little or no effect on corrected visual acuity — a mean difference of 0.00 logMAR. On side effects the picture is thin: nine trials with 333 participants reported on adverse events and three described events occurring, with reported effects from the filtering lenses including increased depressive symptoms, headache, discomfort wearing the glasses and lower mood. That is low-certainty, inconsistently reported evidence. It is mentioned here only because a product sold as risk-free should still have its reported harms stated.

Should I stop wearing mine?

That is your call and there is no medical reason in this review to stop. The lenses did not worsen corrected vision in the one trial that measured it well. What the review changes is what you should expect from them, and what you should pay for that expectation. If they are comfortable and you like them, nothing here argues against wearing them. If you bought them specifically to protect your retina, that is the claim with no trials behind it, and it is worth knowing before you buy the next pair.

What does the evidence actually support for screen-related eye discomfort?

The Cochrane review cannot tell you, because its scope was narrow by design: filtering lenses against non-filtering lenses and nothing else. The American Academy of Ophthalmology does address it, and says plainly that “digital eye discomfort is not caused by blue light” but is instead linked to “device misuse or overuse.” The mechanism it names is drying rather than light: “we blink less when we stare at our devices. Normally, we blink around 15 times per minute – but this ‘blink rate’ can be cut in half when staring at screens.” On sleep specifically it advises that “you don’t need to spend extra money on blue light glasses to improve sleep – simply decrease evening screen time and set devices to night mode.” If your symptoms are burning, grittiness or watering rather than fatigue, dry eye is a separate condition with its own trial literature and is worth raising with an optometrist rather than solving with a purchase.

Sources

• Singh S, Keller PR, Busija L, McMillan P, Makrai E, Lawrenson JG, Hull CC, Downie LE. Blue-light filtering spectacle lenses for visual performance, sleep, and macular health in adults. Cochrane Database of Systematic Reviews 2023 Aug 18;8(8):CD013244. PMID 37593770. DOI 10.1002/14651858.CD013244.pub2. 17 RCTs included; sample sizes 5 to 156 participants; follow-up under one day to five weeks. Verbatim: the reviewers “did not perform meta-analyses for any of the outcome measures, due to lack of available quantitative data, heterogenous study populations, and differences in intervention follow-up periods”; the lenses “may not attenuate symptoms of eye strain with computer use, over a short-term follow-up period”; “we do not know if blue-light filtering spectacle lenses are equivalent or superior to non-blue-light filtering spectacle lenses with respect to sleep quality”; and “there was no evidence from RCT publications relating to the outcomes of contrast sensitivity, colour discrimination, discomfort glare, macular health, serum melatonin levels, or overall patient visual satisfaction.” Visual fatigue mean difference 9.76 units, 95% CI -33.95 to 53.47, 120 participants. Best-corrected visual acuity mean difference 0.00 logMAR, 95% CI -0.02 to 0.02, one study, 156 participants, moderate-certainty evidence. Sleep quality: six RCTs, 148 participants, three reporting improvement and three reporting no significant difference. Risk of bias: none of the studies at low risk across all seven Cochrane RoB 1 domains, 65% at high risk of detection bias, 59% at high risk of performance bias, 35% pre-registered. Adverse events: nine RCTs, 333 participants, low-certainty evidence. Checked 23 August 2026. https://pubmed.ncbi.nlm.nih.gov/37593770/
• American Academy of Ophthalmology, “Should You Be Worried About Blue Light?” (patient education, aao.org). Cited as the position of the professional body rather than as trial evidence, and quoted verbatim: “Blue light from computers will not lead to eye disease”; “the small amount of blue light coming from computer screens has never been shown to harm your eyes”; “so far, the evidence shows no meaningful link between blue light and damage to human retinas or age-related macular degeneration”; “digital eye discomfort is not caused by blue light”; the symptoms after prolonged screen time are “linked not to blue light, but to device misuse or overuse”; and on the mechanism, “we blink less when we stare at our devices. Normally, we blink around 15 times per minute – but this ‘blink rate’ can be cut in half when staring at screens.” Also, on the product itself: “You don’t need to spend extra money on blue light glasses to improve sleep – simply decrease evening screen time and set devices to night mode.” Every quoted sentence was checked against the page source rather than a summary. Checked 23 August 2026. https://www.aao.org/eye-health/tips-prevention/blue-light-digital-eye-strain
• Cochrane search methods for the above review, quoted from the same record: CENTRAL (containing the Cochrane Eyes and Vision Trials Register; 2022, Issue 3), Ovid MEDLINE, Ovid Embase, LILACS, the ISRCTN registry, ClinicalTrials.gov and WHO ICTRP, “with no date or language restrictions,” last searched 22 March 2022. Cited because the absence of trials on macular health is only meaningful if the search was broad, and this records how broad it was. Checked 23 August 2026. https://doi.org/10.1002/14651858.CD013244.pub2