Does Screen Time Cause Myopia in Children?

Every few months a headline announces that screens are ruining children's eyesight. The rise in short-sightedness is real and reasonably well measured. The link to screens specifically is neither, and the gap between those two facts is where most of the advice goes wrong.

Here is what has actually been tested, and what follows from it.

What is actually rising

Myopia is the clinical name for short-sightedness: distant objects blur while near ones stay sharp, because the eyeball has grown slightly too long front to back. It usually starts in primary school and progresses until the late teens.

That it is becoming more common is not in dispute. A 2024 systematic review in the British Journal of Ophthalmology pooled 276 studies covering 5.4 million children and adolescents across 50 countries. Global prevalence rose from 24.3 per cent in 1990 to 35.8 per cent in 2023.

Two caveats the headlines tend to drop. The 1990 estimate carries a confidence interval running from 15 to 33 per cent, so the size of the rise is a good deal less precise than two tidy percentages make it sound. And the pooled studies measured refraction in different ways, some using the pupil-dilating drops that stop a child's own focusing muscles inflating the reading and some not, which moves the number by itself.

Prevalence is also very uneven. The same review put East Asia at 35.2 per cent and adolescents specifically at 47 per cent. Whatever is driving this is not distributed the way global smartphone ownership is.

The screen evidence does not agree with itself

Three serious meta-analyses have asked whether screen time is associated with myopia. They reached three different answers, which is the single most important thing to know about this literature.

A 2020 systematic review in Ophthalmic and Physiological Optics covering 49,789 children aged 3 to 19 pooled five studies and found an odds ratio of 1.02, with a confidence interval from 0.96 to 1.08. That is no association at all. Seven of the fifteen studies it reviewed individually did find one, which tells you how much the answer depends on which studies you pool.

A 2021 meta-analysis in The Lancet Digital Health found an odds ratio of 1.26 for smart device screen time, rising to 1.77 when computer use was included. It is the study most often cited as proof. It also contains a sentence almost nobody quotes: no association was observed when only the prospective studies were pooled. In other words, the link shows up when you ask children about their habits and their eyes on the same day, and disappears when you follow them forward in time.

A 2025 dose-response meta-analysis in JAMA Network Open, the largest of the three at 45 studies and 335,524 participants, found an odds ratio of 1.21 for each additional daily hour, with risk climbing steeply between one and four hours a day and then flattening. A clean dose-response curve is the strongest thing on the pro-screens-matter side of this argument.

What all three share is the problem. Every one is built on observational studies using self-reported or parent-reported screen time. None is randomised. And screen time is an almost perfect proxy for time spent indoors, which turns out to matter.

The lockdown experiment pointed away from screens

The closest thing to a natural experiment happened in 2020, and its result is the opposite of the one it is usually cited for.

Researchers in Feicheng, China had been photoscreening every schoolchild in ten schools annually since 2015. Their 2021 report in JAMA Ophthalmology covered 123,535 children aged 6 to 13 and found a sharp myopic shift after home confinement: a mean change of about 0.32 dioptres among six-year-olds, with prevalence at that age jumping from 5.7 per cent to 21.5 per cent.

That is the number that travelled. The detail that did not: children in grades one and two had one hour a day of online instruction, while grades three to six had two and a half. The largest shift by far was in the youngest children, the ones with the least screen exposure. The authors' own conclusion was that it was unlikely intense screen time or near work had caused it. What had changed for every age group equally was going outside.

Then there is the follow-up, which almost never gets cited at all. A 2023 report on the same cohort, by then covering 325,443 children, found that 2021 prevalence was similar to 2019 in every age band. The spike substantially reversed once children went back outdoors. Because the screening used non-dilated photorefraction, part of the 2020 shift was plausibly a temporary focusing spasm from months of close work rather than a permanently longer eye, and the study had no axial length data to tell the two apart.

What has actually been tested is going outside

Only one environmental factor here has randomised evidence behind it, and it is not screens.

The Guangzhou outdoor activity trial, published in JAMA in 2015, randomised twelve primary schools and 1,903 six-year-olds to an extra 40-minute outdoor class each school day, or to normal schooling, for three years. Cumulative myopia incidence was 30.4 per cent in the intervention group against 39.5 per cent in the control: a difference of 9.1 percentage points, or roughly a quarter fewer children becoming short-sighted.

A Taiwanese trial published in Ophthalmology in 2018 did something the others did not and fitted children with light meters rather than relying on diaries. Children encouraged outdoors showed less myopic shift and slower eye growth, and the authors noted that strong sunlight did not appear to be necessary, so shade counted.

A meta-analysis in Acta Ophthalmologica pooling four trials, eight cohort studies and thirteen cross-sectional studies put the risk ratio for myopia onset at 0.54 in the trials, and estimated that roughly 76 additional minutes outdoors a day corresponded to halving the incidence. Those same authors were careful to say the protective effect was likely overestimated, since most of the underlying outdoor-time data came from questionnaires.

Onset and progression are not the same thing

This is the distinction that decides what any of it is good for, and it is routinely collapsed.

Going outside delays myopia starting. It does very little about myopia that has already started. In the Guangzhou trial the difference in how much myopia actually developed over three years was 0.17 dioptres, which is clinically close to nothing. The Acta Ophthalmologica meta-analysis found no dose-response relationship at all for progression in eyes that were already short-sighted. The Taiwanese trial disagreed and did find a benefit in myopic children, so this is not fully settled.

What is settled is the gap in the evidence. The Cochrane living review of myopia control, covering 64 randomised trials and 11,617 children, records that no included study tested an environmental intervention and reported progression in children who were already myopic. Nobody has run the trial.

So outdoor time is prevention. If your child is already in glasses, it is not the treatment, and anyone selling it to you as one is ahead of the evidence.

What the eye bodies actually recommend

They are strikingly consistent, and none of them says what you would expect.

The American Academy of Ophthalmology's own guidance for parents states plainly that the Academy "does not have specific recommendations for amount of screen time for children". Its 2021 clinical statement puts the reason in one sentence: studies of near work and myopia have produced inconsistent results, whereas the relationship between outdoor time and reduced onset is established more firmly.

The UK College of Optometrists' evidence review reaches the same split verdict: good evidence that time outdoors works, and very limited evidence that reducing near work does.

Where a number is given, it comes from the International Myopia Institute, which recommends at least 80 to 120 minutes a day outdoors for schoolchildren.

Two claims worth correcting while we are here. The widely repeated line that the World Health Organization recommends 90 minutes outdoors a day could not be traced to any WHO document in this research; the WHO pages that discuss myopia recommend outdoor time and less near work without attaching a figure to either. And blue light does not cause myopia. The AAO's own parent page states that it does not recommend blue light glasses, which is a separate question we have looked at directly.

If your child is already short-sighted

Here the evidence gets much better, because this is the part that pharmaceutical and lens companies fund trials on. The Cochrane review above compared the options over two years against a control group that progressed about 1.02 dioptres:

• High-dose atropine drops: the largest effect, around 1.26 dioptres less progression, with pupil dilation and loss of near focus as the trade-off.
• Low-dose atropine: around 0.24 dioptres.
• Orthokeratology, rigid lenses worn overnight to reshape the cornea: about 0.28 mm less eye growth.
• Multifocal soft contact lenses: around 0.30 dioptres.
• Multifocal spectacles: around 0.19 dioptres.
• Deliberately under-correcting a child's glasses: no benefit whatsoever. This was once common advice and the trials are clear that it does not work.

Be careful with the 0.01 per cent atropine that gets discussed most in parent forums, because four good trials produced four different answers. The Hong Kong LAMP trial found it the weakest of three concentrations. The CHAMP trial found 0.01 per cent met all its endpoints while the stronger 0.02 per cent failed its primary one, a result that is not dose-ordered and should lower confidence in the exact numbers. A separate US trial of 187 children found no effect at all. The Irish MOSAIC trial missed its primary endpoint, with a significant result in White children and none in non-White children. This is a live question, not a settled dose.

The newer lenslet spectacle lenses look better. The DIMS lens trial in 183 Hong Kong children reported roughly half the progression of ordinary lenses over two years, and an independent US trial of 135 children completed in 2026 found a similar advantage for a highly aspherical lenslet design, though several of its authors are affiliated with the manufacturer. Orthokeratology works too, and carries a real if uncommon risk: microbial keratitis at roughly five cases per 10,000 patient-years, which the AAO describes as potentially blinding. Low odds, serious worst case, and a decision for an optometrist rather than an article.

What to actually do

The practical version is shorter than the research behind it.

1. Get the outdoor time up. Two hours a day is the working benchmark. It does not need to be sport, and on the Taiwanese evidence it does not need to be bright sunshine. Walking to school, being in the playground and sitting outside all count.
2. Book an eye test rather than watching for symptoms. Children rarely report blurred distance vision, because they have no comparison. Squinting, sitting very close to a screen or television, headaches, or a teacher mentioning the board are all worth acting on, but a routine test finds it earlier.
3. Treat screen limits as a separate decision. There are good reasons to cap a child's screen time. Sleep, attention and what is being displaced are all better supported than eyesight, and our screen time by age guide works through them by stage.
4. Do not bother with blue light glasses for this. No professional body recommends them, and no evidence links blue light to eye growth.
5. Ignore the 20-20-20 rule as a myopia measure. It is aimed at transient eye strain and there is no trial evidence that it changes myopia either way.

If any of this turns into a rule your household has to fight about, the parents hub covers the negotiating half rather than the clinical half.

The honest summary

Short-sightedness in children is rising, screens are associated with it in some studies and not others, and no one has demonstrated that screens cause it independently of the fact that a child on a screen is a child indoors. The one thing with randomised evidence behind it is going outside, and even that mostly delays myopia rather than slowing it once it arrives.

None of this is medical advice, and none of it substitutes for an eye examination. If your child already wears glasses and their prescription is climbing quickly year on year, that is a conversation to have with an optometrist about myopia control, where the evidence is genuinely much stronger than anything on this page about screens.