Do Blue Light Glasses Actually Work?

Blue light glasses are sold on two separate promises. One is that they relieve the tired, gritty, aching eyes you get after a long day at a screen. The other is that they protect your retina from damage, or help you sleep. The evidence behind those promises is not the same, and neither is as strong as the shelf label.

The most useful place to start is not a study. It is the measurement of what the lenses actually do.

The lenses block much less blue light than you would guess

Researchers in Hong Kong put five commercial blue-light-filtering lenses through a spectrophotometer and published the results in PLOS ONE in 2017. The lenses transmitted between 74.9 and 90.3 per cent of blue light. That is, they blocked somewhere between a tenth and a quarter of it.

The same paper calculated what that removal would do to the thing people buy the glasses for. Phototoxicity, the retinal-damage measure, came down by 10.6 to 23.6 per cent. Melatonin suppression, the sleep measure, came down by 5.8 to 15.0 per cent.

Marketing percentages are usually much higher because they quote peak blockage at a single narrow wavelength rather than across the blue band. Both numbers can be true at once. Only one of them describes what reaches your eye.

Hold on to the melatonin figure in particular. A product that removes about a tenth of the relevant signal is being sold against a body clock that, as you will see below, needed four hours of screen exposure to shift.

For eye strain, the trials have not found a benefit

This is the claim on most packaging, and it is the one with the clearest answer.

A Cochrane review published in 2023 pooled every randomised trial of blue-light-filtering spectacle lenses in adults: 17 trials, 619 participants in total, individual trials ranging from 5 people to 156, and follow-up ranging from less than a day to five weeks.

The review could not perform a meta-analysis on any outcome, which tells you most of what you need to know about the state of the literature. The trials were too different from one another and too many reported results only as p-values or in non-numeric form. Its findings, with its own certainty ratings:

• Visual fatigue: there may be no difference over less than a week of use. The single trial with usable numbers, 120 participants, produced an estimate that actually pointed towards the control lenses and was nowhere near significant. Low certainty.
• Visual acuity: probably no effect. Moderate certainty, the highest rating anything in the review received.
• Critical flicker fusion, a measure of visual processing: likely little to no difference. Low certainty.
• Macular health, serum melatonin, contrast sensitivity, colour discrimination, glare discomfort: no randomised evidence at all. Not one of the 17 trials measured them.

The review also recorded adverse events as infrequent but real, including headache, discomfort and, in some trials, lower mood.

Be careful how you read this, because both sides of the argument overstate it. "Cochrane proved blue light glasses do not work" is not what the review says. It says the evidence is low or very low certainty, and that for several of the biggest claims there is no trial evidence to assess. Absence of good evidence is a weaker thing than evidence of absence, and it is also the honest description of what has been published.

What has not happened, in eight years of reviews, is a systematic review finding that the lenses relieve eye strain. An earlier review in 2017 found only three eligible studies and 136 participants. A 2026 narrative review covering literature to March 2025 reached the same conclusion for visual performance and called the eye-strain evidence inconclusive.

The retinal damage claim is the weakest part of the story

The idea that screen light is slowly burning out your retina traces largely to one 2018 laboratory study at the University of Toledo, which got enormous coverage.

That study applied retinal molecules to cultured cells in a dish. It was not a study of human eyes looking at screens. The university's own press release quotes one of the authors saying that the amount of light coming out of a phone "is not great but seems tolerable", which is not how the headlines rendered it.

The measured version comes from the European Commission's scientific committee on emerging health risks, which assessed LED lighting and screens in 2019. Its finding: the public does not risk eye injuries from LED screens in normal use, and screen radiance is less than 10 per cent of the maximum that would still be within safe limits for protecting the retina. The same committee did not give screens a clean bill of health on sleep, where it found some evidence of circadian effects. The two questions are separate and get answered differently.

The American Academy of Ophthalmology's position, published in March 2021, is blunter: there is no scientific evidence that light from computer screens damages the eyes, and the Academy does not recommend any special eyewear for computer use.

Sleep is the one place the question is still open

Here the picture genuinely is unresolved, and anyone telling you otherwise in either direction is ahead of the data.

Start with what is solid. A 2015 study in PNAS had twelve adults live in a monitored sleep unit for two weeks, reading either a light-emitting e-reader or a printed book for four hours before bed on five consecutive evenings. The screen condition suppressed melatonin by 55 per cent, delayed circadian timing by more than 90 minutes, added about ten minutes to how long it took to fall asleep and left people less alert the next morning. Light at the eye measured about 32 lux, which is ordinary dim room lighting rather than anything exotic.

So evening screen light does something measurable. The question is whether a tinted lens removes enough of it to matter, and there the evidence thins out immediately.

Cochrane found six trials of sleep outcomes covering 148 people between them. Three reported an improvement and three reported no difference, which is why the review rated the certainty very low and declined to say whether the lenses help. A 2025 meta-analysis that pooled the three double-blind crossover trials measuring sleep by wrist accelerometer found nothing significant, but pooled only 49 people in total, with a confidence interval on total sleep time running from 35 minutes worse to 53 minutes better. That is not a finding of no effect. It is a question nobody has funded a large enough trial to answer.

There is also a product mismatch hiding in this literature. The eyewear with a plausible mechanism is deep amber and wraparound, worn for hours before bed, usually tested in people with sleep problems. The product sold as blue light glasses is near-clear, worn at a desk in the middle of the day, and removes something like a tenth of the melatonin-relevant signal. The daytime product is being sold on the night-time product's weak evidence.

Night mode on your phone does not do it either

The software version of the same idea has been tested more directly, and it failed.

A trial of 167 young adults published in Sleep Health randomised people to an hour of iPhone use before bed with Night Shift on its warmest setting, the same hour with Night Shift off, or no phone at all, for seven consecutive nights, with sleep measured by wrist accelerometer rather than by asking afterwards.

Nothing moved. Total sleep came out at 400.8 minutes with no phone, 404.8 with an ordinary screen and 406.9 with Night Shift, differences far smaller than the measurement can resolve. Sleep onset latency ran 11.5, 11.9 and 12.2 minutes across the three groups, and sleep efficiency stayed within one percentage point. In an exploratory analysis of the better sleepers, the group that did best was the one with no phone at all, not the one with the filter.

The melatonin work explains why. When researchers at Rensselaer Polytechnic Institute had twelve adults use iPads from 11pm to 1am under different Night Shift settings, every lighting condition suppressed melatonin significantly and the colour settings did not differ meaningfully from one another. Their conclusion was that changing screen colour alone is not enough and brightness has to come down too.

If the goal is sleep, distance from the phone does more than any filter on it, and the rest of the sleep evidence is more mixed than the blue-light story suggests.

Two regulators have already ruled on this

Unusually for a wellness product, the advertising claims have been tested by someone with the power to stop them.

In May 2017 the UK's General Optical Council fined Boots Opticians 40,000 pounds and found its fitness to practise impaired over a newspaper advertisement claiming that blue light from televisions, smartphones and energy-saving bulbs damaged retinal cells over time, and that its lenses protected against it. The claims were found misleading and unsubstantiated. The lenses cost an extra 20 pounds a pair.

In July 2022 the Advertising Standards Authority upheld a complaint against Gamer Advantage over claims that its glasses were clinically proven to attack blue light and would stop the suppression of melatonin. The ASA found the evidence inadequate and noted there was little scientific evidence that blue-light-filtering lenses improved sleep quality. The evidence the company submitted included an unpublished master's thesis and two conference posters, one of them with ten participants and stating that causality could not be established.

The UK College of Optometrists' position statement, published in November 2018 and reaffirmed after the Cochrane review, says the best available evidence does not support these lenses in the general population for visual performance, eye fatigue, sleep quality or macular health.

What actually helps tired eyes

The mechanism behind screen eye strain is boring, which is probably why nobody sells it.

Looking at a screen is a sustained near-focus task in which you blink less, and blink incompletely, so the tear film dries out. That is a plumbing problem, not a spectrum problem, and the things that help follow from it. A 2026 review in Frontiers in Public Health surveyed what has been tested and found, at low certainty across the board:

1. Viewing distance and lighting do the most. Around 60 centimetres from the screen, with sensible ambient light, reduced symptoms most consistently.
2. Breaks help, though not necessarily the famous ones. The review found the standard 20-20-20 rule showed limited benefit, and that longer or self-paced breaks reduced symptoms more. Our full look at the 20-20-20 rule reaches the same place: a free, harmless habit with modest evidence, not a clinical protocol.
3. Artificial tears improve how your eyes feel without changing the objective measures, which is a fair description of what the problem is.
4. Blue-filtering lenses did not improve symptoms or tear film stability compared with control lenses in the study it reviewed.
5. Blink training improved blink behaviour and dryness symptoms in one small study of adolescents.

Turning the screen brightness down, and warming the colour at night, are still worth doing. Not because of blue light exactly, but because total light and how long you sit in it are the parts that measurably matter.

So should you buy a pair

If you already own some, there is no reason to throw them away. They are optically harmless, the trials found adverse events infrequent, and if wearing them makes you take more breaks then the ritual is doing something the lens is not.

If you are about to spend money on them for eye strain, the evidence does not support it, two professional bodies say so in writing, and two UK regulators have penalised advertisers for claiming otherwise. Spend the money on moving the monitor back and getting the room lighting right.

If you are buying them for sleep, buy the right product or none. Clear desk lenses are not what the sleep trials tested, and even the amber ones rest on an evidence base of a few dozen people.

When it is not eye strain

Screen-related discomfort should ease within an hour or so of stopping, and it should not come with anything alarming.

Persistent eye pain, sudden changes in vision, double vision, flashes, floaters, or headaches that do not track your screen use are not digital eye strain and no lens will fix them. They are reasons to see an optometrist or a doctor. An uncorrected prescription, including a small one you have never been tested for, is also one of the most common causes of screen headaches, and it is the one thing on this page that an eye test settles in twenty minutes.

Nothing here is medical advice. If your child is the one with the eye problem, the separate question of screens and short-sightedness has a very different evidence base behind it.