Amblyopia Treatment Options: What PEDIG Trials Show

Amblyopia - most families know it as lazy eye - is one of the few childhood vision conditions with a genuinely large body of randomized-trial evidence behind it. Much of that evidence comes from the Pediatric Eye Disease Investigator Group, or PEDIG, a network of clinical sites that has been running amblyopia trials since the late 1990s.

That evidence answers the questions parents actually ask: Do we have to patch all day? Are drops as good as a patch? Is my nine-year-old too late? Do the game apps and headsets work? Here is what the trials show, what they do not show, and where in-office therapy honestly fits.

Key takeaways

• Amblyopia affects roughly 1.4% of people worldwide and about 2.4% in North America (Fu et al., 2020).
• Glasses alone do a lot: 77% of children with untreated anisometropic amblyopia gained 2 or more lines on optical correction alone, and 27% resolved completely (Cotter et al., 2006).
• For moderate amblyopia, 2 hours of daily patching worked as well as 6 (Repka et al., 2003). For severe amblyopia, 6 hours worked as well as full time (Holmes et al., 2003).
• Atropine drops performed close to patching (2.84 versus 3.16 lines) and are a legitimate alternative (PEDIG, 2002).
• Earlier is better, but older is not hopeless: 53% of 7 to 12-year-olds responded to patching with near activities plus daily atropine, versus 25% on glasses alone (Scheiman et al., 2005).
• Binocular games and headsets are promising but not yet a replacement for standard treatment (AAO, Pineles et al., 2020).

What amblyopia is, and how common it is

Amblyopia is reduced vision in an eye that is structurally healthy. In early childhood the visual brain learns to process input from each eye. If one eye delivers a blurrier or misaligned image - because of a large difference in prescription between the eyes, a big uncorrected prescription in both, an eye turn, or an obstruction such as a cataract or droopy lid - the brain favors the clearer eye and the other never develops normal acuity.

Pooling 60 studies and more than 1.8 million people, one meta-analysis put global amblyopia prevalence at 1.44%, higher in North America (2.41%) and Europe (2.90%) (Fu et al., 2020). In US preschoolers aged 30 to 72 months, the Multi-ethnic Pediatric Eye Disease Study found amblyopia in 1.5% to 2.6% of children and reported that 78% of cases were attributable to refractive error (MEPEDS, 2008). That last number matters more than it looks: most amblyopia starts with a prescription nobody knew about.

Acuity is not the only thing affected. Depth perception is usually the bigger everyday casualty - impaired stereopsis is the most common deficit in amblyopia under normal two-eyed viewing, and it can affect motor coordination, sport and mobility (Levi et al., 2015). We go into what that looks like day to day on our lazy eye page.

Step one is almost always glasses

The most under-appreciated finding in the whole PEDIG program is how much amblyopia improves on glasses alone. In a study of 84 children aged 3 to under 7 with untreated anisometropic amblyopia, optical correction by itself improved acuity by 2 or more lines in 77% of children and resolved the amblyopia entirely in 27%. Improvement kept accumulating for up to 30 weeks (Cotter et al., 2006).

The American Academy of Ophthalmology’s Amblyopia Preferred Practice Pattern states it plainly: “Treatment of refractive error alone is the initial step in care of children 0 through 17 years of age with amblyopia” (AAO PPP, 2023). So if your child has just been prescribed glasses for amblyopia, give them a real trial period before assuming a patch is coming. Several months of consistent wear may be the whole treatment.

Patching: how many hours actually helps

Patching covers the stronger eye so the amblyopic eye is forced to work. For decades the instinct was more hours, all day, no exceptions. PEDIG tested that instinct.

Moderate amblyopia: two hours is usually enough

In 189 children under 7 with moderate amblyopia (20/40 to 20/80), 2 hours and 6 hours of daily patching produced exactly the same mean improvement - 2.40 lines at 4 months - with a 62% success rate in each group (Repka et al., 2003). Each group also did an hour of near activities while patched.

Severe amblyopia: six hours, not the whole day

In 175 children under 7 with severe amblyopia (20/100 to 20/400), 6 hours of prescribed daily patching improved acuity 4.8 lines, versus 4.7 lines with full-time patching (Holmes et al., 2003).

The AAO PPP reflects both results: 2 hours “produces an improvement in VA that is similar in magnitude to the improvement produced by 6 hours” for moderate amblyopia, while “for amblyopia worse than 20/80, 6 or more hours of patching is often prescribed” (AAO PPP, 2023).

This is one of the kindest findings in pediatric eye care. A dose a family can actually complete beats a heroic dose they abandon. We wrote more about how patching is used alongside active therapy in The Role of Eye Patches in Vision Therapy.

Atropine drops: a real alternative

Atropine penalization uses a daily drop to blur the stronger eye’s near vision, shifting work to the amblyopic eye without a patch on the face. In 419 children under 7 with moderate amblyopia, acuity improved 3.16 lines with patching and 2.84 lines with atropine at 6 months, with success rates of 79% and 74% (PEDIG, 2002). Parents in that trial found atropine somewhat easier to accept.

Followed to age 15, 147 of those participants had a mean amblyopic-eye acuity of about 20/25, with 59.9% seeing 20/25 or better - and outcomes did not differ by which treatment they were originally assigned. The same follow-up is honest about the ceiling: 48.3% still had a difference of 2 lines or more between their eyes (Repka et al., 2014). Treatment usually improves amblyopia substantially. It does not always erase it.

Does age close the door?

Earlier is better, and the trials say so clearly. A meta-analysis of individual data from four PEDIG trials found children aged 7 to under 13 were significantly less responsive than children aged 3 to 7, with no meaningful difference between 3 to 5 and 5 to 7 year-olds for moderate amblyopia (Holmes et al., 2011).

But “less responsive” is not “unresponsive.” In a trial of 507 patients aged 7 to 17, 53% of the 7 to 12-year-olds responded to the treatment arm - 2 to 6 hours a day of patching combined with near visual activities, plus daily atropine at that age - compared with 25% on optical correction alone. Among 13 to 17-year-olds the overall difference was small - but in teenagers who had never been treated before, 47% responded versus 20% (Scheiman et al., 2005). The AAO PPP agrees that “patching should be considered for older children and teenagers, particularly if they have not previously been treated” (AAO PPP, 2023).

If you have been told your child aged out of treatment, that is worth a second opinion. Adults carrying a long-standing eye turn or double vision have a separate set of options, which we cover in Adult Strabismus and Double Vision.

Binocular and dichoptic games: promising, not proven

Dichoptic treatment shows each eye a different part of an image or game, so the two eyes must cooperate to see the whole picture. It is the most interesting new direction in amblyopia care - and the evidence is genuinely mixed, so we will not simplify it.

• PEDIG iPad game, ages 5 to 12. In 385 children, the binocular game improved acuity 1.05 lines versus 1.35 lines with 2 hours of daily patching. Of the 176 children in the game group whose usage logs were available, only 22.2% completed more than 75% of the prescribed play (Holmes et al., 2016).
• PEDIG iPad game, teens. In 100 teenagers, the game gained 3.5 letters versus 6.5 with patching, and only 13% completed more than 75% of treatment (Manh et al., 2018).
• BRAVO, ages 7 and up. Against a placebo game, the active binocular game showed no benefit (0.06 versus 0.07 logMAR) (Gao et al., 2018).
• Dichoptic digital therapeutic, ages 4 to 7. A pivotal trial of 105 children found 1.8 lines of improvement with the therapeutic plus glasses versus 0.8 lines with glasses alone at 12 weeks (Xiao et al., 2022). The comparator was glasses, not patching - so it does not show superiority to a patch.
• Pooled evidence. A 2026 systematic review of 12 randomized trials and 1,034 children found no significant difference between dichoptic therapy and patching for visual acuity (mean difference -0.01, 95% CI -0.05 to 0.03), stereoacuity, compliance or adverse events (Halabi & Aldhahwani, 2026).

Reviewing 20 studies in a literature search that closed in April 2019 - so before the two most recent entries above - the American Academy of Ophthalmology concluded there was no level I evidence supporting binocular treatment as a substitute for patching or optical treatment (Pineles et al., 2020), and the 2023 Preferred Practice Pattern still says “the evidence remains limited to recommend binocular therapy for treatment” (AAO PPP, 2023).

Adherence was poor in both PEDIG game trials, and the investigators were explicit that they could not tell whether the weak response reflected too little use or too little effect. Engagement is probably not the whole story, though: in the placebo-controlled BRAVO trial, 64% of the active group and 83% of the placebo group completed more than a quarter of the prescribed play, and the active game still did no better than placebo.

If you are curious how these tools are being used in clinics, we cover that in Virtual Reality and Vision Therapy.

Where in-office vision therapy fits - and the AAO caveat

We should be straightforward here, because it is our own field. The AAO Preferred Practice Pattern reviews vision exercises for amblyopia and states there is “no convincing evidence for treatment success with these activities for amblyopia” (AAO PPP, 2023). Anyone who tells you office-based therapy is the proven way to raise amblyopic acuity is going beyond the evidence.

Our clinical perspective at Vision & Learning Center: we do not offer vision therapy as a substitute for glasses, patching or atropine. We use it for the problems that sit alongside amblyopia and are not addressed by acuity treatment - eye teaming and depth perception once the two eyes can meaningfully work together, eye-movement control for reading, focusing stamina, and the visual-motor skills that lag when a child has spent years functioning on one eye. We also use active near tasks during patching, because that is how PEDIG’s patching arms were actually run. When amblyopia comes with a constant eye turn, we co-manage with pediatric ophthalmology rather than work around it. You can read what that looks like from a therapist’s chair in Amblyopia in the Eyes of the Therapist, and more about our approach on our amblyopia and strabismus page.

The part treatment plans often miss: how it feels

A patch is worn on a child’s face, at school, in front of everyone. A 2025 systematic review of 25 studies published between 1999 and 2021 found that both children and their guardians can experience distress from amblyopia treatment, and that occlusive patching - and patching during the school years in particular - raises the likelihood of distress most. The authors suggested earlier diagnosis and considering atropine as a first-line option to protect wellbeing while still treating effectively (Haine et al., 2025).

This is not a reason to skip treatment. It is a reason to choose the dose and the method with the child in the room, to patch after school rather than during it where the clinical picture allows, and to take teasing seriously when a parent reports it. We wrote about that side of care in The Social Implications of Eye Turns and Patching.

Catching it early is still the biggest lever

Amblyopia rarely announces itself - a child with one good eye reports nothing wrong - so detection is the real bottleneck. The US Preventive Services Task Force recommends vision screening at least once between ages 3 and 5 to detect amblyopia or its risk factors (USPSTF, 2017). The American Optometric Association’s pediatric guideline goes further, recommending a comprehensive eye examination at 6 to 12 months, at least once between 3 and 5 years, and before first grade and annually after that (AOA, 2017). A screening is a filter; an examination measures refraction, alignment, eye teaming and depth perception directly.

If your child has an eye that turns, squints or closes in bright light, has failed a screening, or has a family history of amblyopia or strabismus - and especially if you have been told to simply wait and see - we would rather measure it. Schedule an evaluation at our Boynton Beach or Boca Raton office, or read more about what a comprehensive developmental vision evaluation includes.