Industries / Health / Healthcare / Optometry
Slowing short sight in children
Myopia control uses treatments and habits to slow a child's eye from growing too long, which causes myopia, or short sight, and later the eye diseases that high myopia brings.
7
Milestones achieved, from 2006 to 2024.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2006
Eye drops that slow myopia
In the ATOM1 trial in Singapore, 400 children aged 6 to 12 put nightly drops of 1 percent atropine, a drug that relaxes the eye's focusing muscle, or dummy drops into one eye. Over two years myopia progressed by 1.20 diopters (D, the unit of lens power) in control eyes and by 0.28 D in treated eyes, and treated eyes barely grew longer.
Lessonthe dose that worked also blurred near vision and widened the pupil, so the next trial tested far weaker drops.
Credit
Wei-Han Chua, Donald Tan and colleagues (Singapore National Eye Centre)
2014
Myopia rebounds after stopping
The ATOM2 trial gave 400 Singapore children 0.5, 0.1 or 0.01 percent atropine for two years, and myopia progressed by 0.30, 0.38 and 0.49 D. In the year after the drops stopped, it progressed by 0.87 D in the 0.5 percent group and by 0.28 D in the 0.01 percent group.
Lessonstronger doses worked better while given but gave back more afterwards, which turned attention to the weakest dose.
Credit
Audrey Chia, Donald Tan and colleagues (Singapore National Eye Centre and Singapore Eye Research Institute)
2015
Time outdoors prevents myopia
A trial in 12 primary schools in Guangzhou, China, with 1,903 six-year-olds added one 40-minute outdoor class to each school day. Over three years 30.4 percent of eligible children in those schools became myopic, against 39.5 percent in schools that changed nothing. A smaller 2013 study in southern Taiwan, which sent children outside at recess, had found new myopia in 8.4 percent against 17.7 percent after one year.
Credit
Mingguang He and colleagues (Guangzhou); Pei-Chang Wu and colleagues (Kaohsiung Chang Gung Memorial Hospital, Taiwan)
2016
Half the world, by 2050
A meta-analysis of 145 studies covering 2.1 million people estimated that 1.4 billion people, 22.9 percent of the world, were myopic in 2000. It projected 4.76 billion, 49.8 percent, by 2050, with 938 million highly myopic (−5.00 D or worse). Later trial reports, such as CHAMP's in 2023, open with this figure.
Credit
Brien Holden, Timothy Fricke and colleagues (Brien Holden Vision Institute, Sydney)
2019
A lens approved for myopia
On 15 November 2019 the FDA approved CooperVision's MiSight, a daily disposable soft contact lens with rings that focus part of the light in front of the retina, for children aged 8 to 12. The FDA called it "the first FDA-approved product to slow the progression of myopia in children". In the three-year trial, refraction changed by 0.51 D with MiSight against 1.24 D with an ordinary lens, 59 percent less.
Credit
CooperVision; Paul Chamberlain and colleagues (trial)
2023
Weak drops fail in America
A US trial by the Pediatric Eye Disease Investigator Group gave 187 children of many backgrounds 0.01 percent atropine or placebo for two years. Myopia progressed by 0.82 D with atropine and 0.80 D with placebo, no difference. The same year the CHAMP trial of Vyluma's drops, in North America and Europe, reported that 0.01 percent atropine slowed progression over three years by 0.24 D against placebo, while 0.02 percent missed its main goal.
Lessonresults in East Asian children did not settle what weak atropine does elsewhere, and two trials of the same dose in the same year disagreed.
Credit
Michael Repka and colleagues (PEDIG); Karla Zadnik and colleagues (CHAMP)
2024
Red-light devices pulled in China
In repeated low-level red-light therapy a child looks into a red laser device for 3 minutes twice a day. In a 2022 Chinese trial eyes grew 0.13 mm in a year with it against 0.38 mm with ordinary glasses. A 2023 case report then described retinal damage after the treatment, and tests at the University of Houston found that laser devices reached US eye-safety limits within seconds, well inside the 3-minute session. China's National Medical Products Administration reclassified the devices as high risk (Class III), and from 1 July 2024 devices without a Class III certificate could no longer be made or sold.
Lessonthe treatment reached many children before anyone outside the makers measured how much light the devices put into the eye.
europepmc.orgeuropepmc.orgpmc.ncbi.nlm.nih.goveuropepmc.orgeuropepmc.orgmyopiaprofile.com
Credit
Yu Jiang, Mingguang He and colleagues, Zhongshan Ophthalmic Center (trial); Lisa Ostrin and Alexander Schill (device tests); Ya Xing Wang, Ningli Wang and Tien Yin Wong (account of the rules)
Watch for
Atropine drops in America
Sydnexis says its 0.01 percent atropine drop SYD-101 works, citing its phase 3 STAR trial, which met its main goal. The European Commission approved it in June 2025 and the UK approved it as Ryjunea on 6 November 2025, but the FDA declined it in October 2025, saying the data did not support its effectiveness. Sydnexis appealed, and on 30 October 2026 an FDA advisory committee discussed the drug, where FDA staff said the overall results suggested "limited to no clinical benefit".
Achieved whenthe FDA approves a low-dose atropine drop for myopia, or an independent trial confirms SYD-101's effect.
Claimed by
Sydnexis, October 2026
A whole city's schools
A trial in 68 Shanghai primary schools with 43,185 children aged 6 to 12 combined at least one hour outdoors each school day with myopia-control glasses. After six months, new myopia was 8.66 percent in intervention schools against 13.50 percent in control schools, and eyes grew 48 percent more slowly.
Achieved whenfull results over at least two years are published in a peer-reviewed journal.
Claimed by
Xiangui He and colleagues, Shanghai Eye Disease Prevention and Treatment Center, at the ARVO meeting, May 2026
Open
01
Registered trials that never reported
A list of every myopia-control trial in children registered on ClinicalTrials.gov, the Chinese Clinical Trial Registry and the other registries in the World Health Organization's trial portal, each matched to its published results or marked unpublished, with the share left unpublished by treatment and by sponsor.
The briefCredit · unclaimed02
Myopia claims checked against labels
A scored public sample of clinic, maker and seller web pages in the US, UK and Australia that advertise myopia control, comparing each claim (the percent slowing, the ages treated, whether the treatment is approved) with the product's approved label and its trial results.
The briefCredit · unclaimed
Open 01 · the brief
Registered trials that never reported
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A list of every myopia-control trial in children registered on ClinicalTrials.gov, the Chinese Clinical Trial Registry and the other registries in the World Health Organization's trial portal, each matched to its published results or marked unpublished, with the share left unpublished by treatment and by sponsor.
Why it is still open
A 2025 study of 49 trials found no sign of publication bias, but it tested only trials already published, using funnel plots, and did not look up registered trials that never reported. No count of unpublished myopia trials was found.
Built with what exists
the ClinicalTrials.gov API, the WHO International Clinical Trials Registry Platform export, the Europe PMC API, and an AI agent that proposes matches between registrations and papers for a person to confirm.
First step
pull all ClinicalTrials.gov records with myopia as the condition and children in the eligibility, and count those marked completed before 2023.
Done looks like
a public table of all trials marked completed before 2023, each marked published or unpublished with a link, plus 50 matches checked by a second person, with the rate of disagreement reported.
Open 02 · the brief
Myopia claims checked against labels
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A scored public sample of clinic, maker and seller web pages in the US, UK and Australia that advertise myopia control, comparing each claim (the percent slowing, the ages treated, whether the treatment is approved) with the product's approved label and its trial results.
Why it is still open
Regulators approve narrow uses, for example MiSight for children aged 8 to 12 at the start of treatment. A 2026 study of 120 patient web pages on myopia treatments scored their reading level and whether they named authors and dates, but not whether their claims were accurate.
Built with what exists
a web crawler, a language model that pulls each claim into fixed fields for a person to check, and the FDA approval records and UK and EU product summaries as the reference.
First step
write the scoring sheet from the MiSight approval and the Ryjunea product summary, and collect the first 50 pages from search results for "myopia control" in each country.
Done looks like
at least 300 pages scored with saved copies, the share making a claim beyond the approved label, and the scoring sheet published so others can repeat it.