Deprivation<\/strong><\/td>\n| Something physically blocks the visual axis \u2014 congenital cataract, corneal opacity, a droopy lid<\/td>\n | Least common, typically the most severe, and by far the most time-critical<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n One caution on the numbers. You will see a tidy split quoted \u2014 around 50% refractive, 19% strabismic, 27% combined, 4% deprivation \u2014 but that comes from a single study population, and a U.S. study of African American and Hispanic\/Latino children attributed 78% of amblyopia to refractive error<\/strong>. The honest summary is “most cases are refractive,” not a precise percentage.<\/p>\nStrabismus<\/strong> is ocular misalignment \u2014 the eyes simply are not lined up. It affects a similar share of children: two large U.S. population studies of kids aged 6 to 72 months found 2.1\u20133.3%<\/strong> (Baltimore) and 2.4\u20133.55%<\/strong> (Multi-Ethnic, Los Angeles), with inward and outward turns each accounting for close to half of cases.<\/p>\n\n- Esotropia<\/strong> \u2014 an eye turns inward, toward the nose. Sometimes driven by uncorrected farsightedness, in which case glasses alone can straighten it.<\/li>\n
- Exotropia<\/strong> \u2014 an eye drifts outward. The most common childhood form is intermittent<\/strong>, showing up when the child is tired, sick, daydreaming, or looking into the distance.<\/li>\n
- Hypertropia<\/strong> \u2014 one eye sits higher than the other. Often the reason behind a persistent head tilt.<\/li>\n<\/ul>\n
The two conditions feed each other. Strabismus can cause amblyopia by suppression, and uncorrected refractive error can cause strabismus. That is why a visible eye turn needs both an alignment check and a refraction check.<\/p>\n ![\"Two-panel](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-1-4.png\") <\/p>\n
\nWhat the sensitive period really means<\/h2>\nHere is the part that gets flattened into “age 3.”<\/p>\n Basic visual wiring is laid down very early \u2014 the most fundamental phase is often described as birth to about three months. After that comes a longer sensitive period<\/strong>, during which the visual cortex stays plastic. Amblyopia can both develop and be reversed inside that window. Plasticity is at its maximum in the first 3 to 5 years<\/strong>, then declines progressively through late childhood, conventionally placed around age 7 to 8<\/strong>.<\/p>\nTwo things get mixed up constantly, and separating them fixes most of the confusion:<\/p>\n \n- Susceptibility<\/strong> \u2014 how likely a child is to newly develop amblyopia. This drops with age.<\/li>\n
- Treatability<\/strong> \u2014 how much a child who already has it can gain. This also drops with age, but it does not fall to zero at any birthday.<\/li>\n<\/ul>\n
What the trials actually found<\/h3>\nThe Pediatric Eye Disease Investigator Group (PEDIG), funded by the National Eye Institute, ran the definitive randomized studies.<\/p>\n ATS3 (2005)<\/strong> enrolled children aged 7 to 17. In the 7-to-12 group (n = 404), 53% of the treated children were responders, versus 25% on glasses alone.<\/strong> That is not a marginal result. In the 13-to-17 group (n = 103)<\/strong> the overall numbers were 25% versus 23% \u2014 but the authors concluded that patching with near activities may improve acuity in teenagers whose amblyopia had never been treated before<\/strong>, while adding little for those who had already completed a full course.<\/p>\nA 2011 pooled analysis<\/strong> of individual patient data from four PEDIG randomized trials shows the gradient more precisely. For severe amblyopia, adjusted mean improvement was:<\/p>\n\n\n\n| Age band<\/th>\n | Mean improvement (severe amblyopia)<\/th>\n<\/tr>\n<\/thead>\n | \n\n| 3 to under 5<\/td>\n | 4.16 lines<\/td>\n<\/tr>\n | \n| 5 to under 7<\/td>\n | 3.60 lines<\/td>\n<\/tr>\n | \n| 7 to under 13<\/td>\n | 1.99 lines<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n Two things to take from that table. Children aged 7 to under 13 were significantly less responsive<\/strong> \u2014 the direction of the evidence is unambiguous that younger is better. And they still gained about two lines<\/strong>, which is a real functional change, not a rounding error.<\/p>\nThere is a smaller surprise in that same analysis: there was no statistically significant difference between the 3-to-under-5 and 5-to-under-7 bands.<\/strong> Even the “before kindergarten or bust” version of the deadline is softer than commonly assumed.<\/p>\nAAO’s own summary of this body of work frames it as a gradient: starting before age 5 yields greater improvement and stability than starting between 7 and 17.<\/p>\n ![\"A](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-2-3.jpg\") <\/p>\n
Photo: Kampus Production \/ Pexels<\/em><\/p>\nSo the sentence I would want a parent to leave with is this one.<\/strong> Earlier is genuinely, measurably better \u2014 and “too late” is not a thing anyone should be told about a school-age child. AAPOS is equally direct about the far end of it: the treatments we have do not work for adults.<\/strong> The window is real. It just closes slowly, over years, rather than slamming at a third birthday.<\/p>\n
\nThe one deadline that really is measured in weeks<\/h2>\nDeprivation is the exception that probably created the deadline myth in the first place \u2014 and the real numbers are far tighter than age 3.<\/p>\n For a dense unilateral congenital cataract<\/strong>, best outcomes are associated with surgery within roughly 6 weeks of birth<\/strong>. For dense bilateral<\/strong> cataract, the equivalent latent period is around 10 weeks<\/strong>. Better visual outcomes generally require surgery within the first three months of life, and severe form deprivation can leave nystagmus even after early surgery.<\/p>\nAnd surgery is not the end of it. The child still needs refractive correction afterward \u2014 a contact lens or aphakic glasses \u2014 and for one-sided cases, patching of the better eye. Removing the cataract clears the path; it does not by itself teach the brain to use the eye.<\/p>\n This is the strongest single argument for the newborn red reflex test<\/strong>, and the reason nobody in this field is content to wait until preschool.<\/p>\n![\"Horizontal](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-3-3.png\") <\/p>\n
\nSigns parents can actually notice at home<\/h2>\nStart with the uncomfortable premise: your child will almost certainly not tell you.<\/strong><\/p>\nA child with one good eye typically has no symptoms and no idea anything is wrong. There is no pain, no obvious behavior change, and school performance is often normal \u2014 because they have never known anything different. AAO makes this point directly in its screening material. “He seems to see fine” is not reassurance. It is the expected presentation.<\/p>\n What you can see instead is behavioral and physical.<\/p>\n \n- Eyes that look crossed, turn out, or don’t focus together.<\/strong><\/li>\n
- An eye that drifts<\/strong> \u2014 especially when your child is tired, sick, daydreaming, or looking off into the distance. That is the classic intermittent exotropia pattern, and because it comes and goes, it is the one families most often talk themselves out of.<\/li>\n
- A persistent head tilt, face turn, or chin-up posture.<\/strong> Children adopt these to line their eyes up for the best available vision or to avoid double vision. If your child routinely tilts their head in an unusual way to look at something, that is worth mentioning to a doctor.<\/li>\n
- Squinting or closing one eye, especially in bright sunlight.<\/strong> AAO notes this is often done to block out a double image from misaligned eyes.<\/li>\n
- Poor depth perception<\/strong> \u2014 clumsiness, trouble catching a ball, missing the last step on the stairs.<\/li>\n
- Sitting very close to screens, holding books very close.<\/strong><\/li>\n
- Eye rubbing, excessive tearing, eye fatigue, light sensitivity.<\/strong><\/li>\n
- A drooping eyelid<\/strong>, which can itself be a deprivation cause if it blocks the visual axis.<\/li>\n
- Nystagmus<\/strong> \u2014 eyes that flutter quickly side to side or up and down.<\/li>\n
- Asymmetric red-eye in flash photos<\/strong> \u2014 a red reflection in only one eye rather than both.<\/li>\n<\/ul>\n
There is one gentle observation you can make at home, and I want to frame it carefully. While your child is looking at something interesting, cover one eye with your hand, then the other, and watch the reaction. A strong, consistent objection to one specific eye being covered<\/strong> \u2014 pushing your hand away, fussing, turning the head \u2014 suggests the uncovered eye is the weaker one. It mirrors what clinicians watch for in children too young to read a chart.<\/p>\nThat is an observation prompt, not a test. A child who tolerates both sides equally can still have amblyopia. Only a professional exam rules it out.<\/p>\n ![\"A](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-4-3.jpg\") <\/p>\n
Photo: Mochi Mochi \/ Pexels<\/em><\/p>\n
\nThe emergency sign: a white pupil in a flash photo<\/h2>\nAlmost everything above belongs in a normal appointment. This one does not.<\/p>\n Leukocoria<\/strong> is a white, grayish-white, or yellow pupil. It is most often first spotted by a parent in a flash photograph<\/strong>, where one eye glows red and the other glows white. The AAP lists a white or grayish-white color in the pupil explicitly as a warning sign.<\/p>\nLeukocoria is the most common and often the first sign of retinoblastoma<\/strong>, a childhood eye cancer, present in approximately 60% of retinoblastoma cases.<\/strong> A white pupillary reflex or a squint are the most common presenting signs, and affected children usually show no other signs of illness at all.<\/strong><\/p>\nIt has other causes \u2014 congenital cataract, Coats disease, persistent fetal vasculature, retinopathy of prematurity, toxocariasis. Every one of them is urgent. Any suspicion of an abnormal red reflex warrants immediate evaluation by an ophthalmologist<\/strong>. Early detection materially improves the chance of saving both the eye and the child’s life.<\/p>\nTwo other situations belong in the same category:<\/p>\n \n- A new, sudden eye turn<\/strong> in a previously straight-eyed child, particularly alongside headache, vomiting, double vision, a droopy lid, or unequal pupils.<\/li>\n
- Any constant eye turn, at any age, including in a newborn.<\/strong> A constant turn is never normal.<\/li>\n<\/ul>\n
Do not save these for the next well-child visit.<\/p>\n
\nCrossed eyes that aren’t: pseudostrabismus<\/h2>\nThere is a common, entirely benign look-alike, and knowing about it cuts both ways.<\/p>\n Pseudostrabismus<\/strong> is the appearance of crossed eyes in a child whose eyes are perfectly aligned. It comes from a wide, flat nasal bridge<\/strong> and prominent epicanthal folds<\/strong> \u2014 small folds of eyelid skin at the inner corner that cover the nasal white of the eye, creating the illusion of an inward turn. Most cases resolve by age 2 to 3<\/strong> as the nasal bridge grows.<\/p>\nThe distinguishing check is the corneal light reflex<\/strong>, and it is the same one your pediatrician does with a penlight:<\/p>\n\n- Pseudostrabismus:<\/strong> the small light reflection sits in the same place in both eyes.<\/strong><\/li>\n
- True strabismus:<\/strong> the reflection sits in a different place in each eye.<\/strong><\/li>\n<\/ul>\n
This is also why a straight-on flash photo can be genuinely useful \u2014 you can look at where that pinpoint reflection lands on each pupil.<\/p>\n And here is the caveat that matters more than the fact itself. Pseudostrabismus is the single biggest reason a real<\/strong> eye turn gets waved away as “she’ll grow out of it.” A child can have pseudostrabismus and<\/strong> true strabismus at the same time. The reassurance has to come from a clinician, never from a relative and never from a photo you squinted at online.<\/p>\n![\"Side-by-side](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-5-2.png\") <\/p>\n
When is crossing normal in a baby?<\/strong> Occasional inward or outward straying in the first couple of months is common while the eyes learn to work together. But inward crossing after 3 months, or outward drifting after 4 months, is not normal<\/strong>, does not typically resolve on its own, and warrants referral. That allowance applies only to intermittent drifting \u2014 never to a constant turn.<\/p>\n
\nThe screening schedule, age by age<\/h2>\nThe AAO and AAPOS, working with the AAP, updated their joint clinical statement on vision screening in 2022<\/strong>. This is the practical map.<\/p>\n![\"A](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-6-4.jpg\") <\/p>\n
Photo: Gustavo Fring \/ Pexels<\/em><\/p>\n\n\n\n| Age<\/th>\n | What should happen<\/th>\n<\/tr>\n<\/thead>\n | \n\nNewborn<\/strong><\/td>\nEye health check and a red reflex test<\/strong> in the newborn nursery. Infants with a family history of retinoblastoma, glaucoma, cataracts, or retinal conditions go straight to an ophthalmologist<\/td>\n<\/tr>\n\n1 month \u2013 4 years<\/strong><\/td>\nOcular health assessed at every routine well-child visit<\/strong>: external inspection, pupils, corneal light reflex, fixation and following<\/td>\n<\/tr>\n\n12 months \u2013 3 years<\/strong><\/td>\nInstrument-based photoscreening or handheld autorefraction may be performed<\/strong>, allowing earlier detection of conditions that lead to amblyopia. Policy is to first attempt it between 12 months and 3 years, then annually until acuity can be tested directly<\/td>\n<\/tr>\n\n3 \u2013 5 years<\/strong><\/td>\nVisual acuity testing as soon as the child cooperates \u2014 generally between 3\u00bd and 4. Formal acuity testing must occur by age 5<\/strong><\/td>\n<\/tr>\n\n5+ years<\/strong><\/td>\nScreening every 1 to 2 years<\/strong> at school checks or well-child visits<\/td>\n<\/tr>\n\nAny age<\/strong><\/td>\n| A failed screening \u2014 or an uncooperative child on a second attempt \u2014 requires a comprehensive exam by an ophthalmologist<\/strong><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n Photoscreening<\/strong> deserves a plain-English note, because it is the piece most parents have never heard of. It is a handheld, camera-like device that reads the reflection off the retina and flags amblyopia risk factors<\/strong> \u2014 significant refractive error, unequal error between the eyes, media opacity, misalignment. It takes seconds, requires almost no cooperation, and does not depend on the child answering anything, which makes it genuinely useful in preverbal, preliterate, or developmentally delayed children.<\/p>\nIts limitation is in that same sentence: it detects risk factors, not amblyopia itself.<\/strong> A pass on an instrument screen is not a comprehensive eye exam.<\/p>\nWhat the acuity numbers mean<\/h3>\nIf your child is handed a symbol chart, these are the referral thresholds AAPOS uses.<\/p>\n \n\n\n| Age<\/th>\n | Refer if worse than<\/th>\n<\/tr>\n<\/thead>\n | \n\n| 3 years (36\u201347 months)<\/td>\n | 20\/50<\/td>\n<\/tr>\n | \n| 4 years (48\u201359 months)<\/td>\n | 20\/40<\/td>\n<\/tr>\n | \n| 5 years and older<\/td>\n | 20\/32 (LEA\/HOTV\/Sloan) or 20\/30 (Snellen)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n A two-line difference between the eyes<\/strong> is a red flag even when both eyes technically pass. Which is exactly why the test is done one eye at a time<\/strong> \u2014 the reason your child is asked to hold a paddle over one eye rather than just read the chart.<\/p>\nAn honest disagreement between two authorities<\/h3>\nThis is worth knowing about, because if you go looking you will find both.<\/p>\n The U.S. Preventive Services Task Force<\/strong> gives a Grade B recommendation to screen at least once in all children aged 3 to 5<\/strong> to detect amblyopia or its risk factors \u2014 meaning it recommends the service, with moderate certainty of moderate net benefit. For children younger than 3<\/strong>, it issues an I statement: the current evidence is insufficient<\/strong> to assess the balance of benefits and harms.<\/p>\nMeanwhile the AAO\/AAPOS\/AAP 2022 policy endorses instrument-based screening from 12 months.<\/strong><\/p>\nThat looks like a contradiction and mostly isn’t. “Insufficient evidence” means the trials have not been done \u2014 not that screening is harmful or useless. The specialty societies weigh mechanistic and observational evidence more heavily and act on it. Worth noting on dates, too: the USPSTF statement is from September 2017 and is currently flagged as under review<\/strong>, so treat it as the standing recommendation rather than a fresh one.<\/p>\nOne more practical change, from a 2025 update<\/strong> in Optometry and Vision Science<\/em>: it recommends annual screening for ages 3 to under 6<\/strong>, endorses LEA Symbols and HOTV letters as the best optotypes for preschoolers, and \u2014 the change I would actually act on \u2014 shortens the rescreening window for a child who couldn’t complete the test from “no later than 6 months” to “no later than 2 weeks.”<\/strong> If your child was untestable, don’t accept “come back in six months.”<\/p>\nAnd the gap this is all trying to close is real. A 2024 analysis reported vision-screening prevalence among U.S. children aged 5 and under hovering near 40% during 2016\u20132020<\/strong>, with fewer than 15%<\/strong> receiving an eye exam from an eye care professional. (An older CDC data brief using 2016\u20132017 survey data reports age-by-age percentages that are widely quoted; I could not retrieve that page directly to verify them, so I am leaving those specific figures out.)<\/p>\n![\"Vertical](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-7-3.png\") <\/p>\n
\nTreatment, and why it is less punishing than you think<\/h2>\nGlasses first \u2014 and often glasses only<\/h3>\nCorrecting the refractive error is first-line, and it is frequently enough on its own.<\/p>\n PEDIG found that 77% of children with anisometropic amblyopia improved by two or more lines within 15 weeks of wearing glasses alone.<\/strong> This is why the standard sequence is glasses first, with a period of refractive adaptation, before anything is added. A substantial number of children never need a patch at all. Accommodative esotropia \u2014 crossing driven by uncorrected farsightedness \u2014 can straighten with glasses alone too.<\/p>\nPatching<\/h3>\nThe mental image most adults carry is an all-day eyepatch for years. The trials say otherwise.<\/p>\n \n- Moderate amblyopia: 2 hours a day of patching is as effective as 6 hours a day<\/strong>, and as full-time patching, in children aged 3 to 7. About 62% reached 20\/30 or gained at least three lines within four months.<\/li>\n
- Severe amblyopia (20\/100 to 20\/400): 6 hours a day is as effective as full-time<\/strong>, with a mean improvement of 4.1 lines.<\/li>\n<\/ul>\n
Shorter regimens paired with near visual activities<\/strong> \u2014 reading, drawing, handheld games under the patch \u2014 are the standard now. Hours must be prescribed and monitored, because over-patching risks reverse amblyopia in the good eye.<\/p>\nAtropine drops<\/h3>\nAtropine 1% goes in the stronger<\/strong> eye. It dilates the pupil and paralyzes accommodation, blurring near vision in the good eye and forcing the amblyopic eye to work.<\/p>\nIt is as effective as patching for moderate amblyopia<\/strong>, with better reported compliance and quality of life \u2014 no visible patch, no daily negotiation. And weekend-only dosing works<\/strong>: a randomized trial of 168 children under 7 found two days a week as effective as daily.<\/p>\nSide effects are usually mild \u2014 light sensitivity from the dilated pupil, and near blur, which persists just under or over 48 hours after a dose. In one long-term series, photochromic spectacles for light sensitivity were needed in 31\u201333% over two years<\/strong>, while progressive lenses for near blur were needed in only about 1%<\/strong>. Reverse amblyopia is a real risk here too, which is why scheduled follow-up is not optional.<\/p>\nSurgery \u2014 for alignment, not for amblyopia<\/h3>\nThis distinction gets lost constantly. Strabismus surgery realigns the eye muscles.<\/strong> AAO states explicitly that surgery does not replace eyeglasses or patching<\/strong>. It corrects how the eyes point; it does not by itself cure the amblyopia, which still needs optical correction and occlusion or penalization.<\/p>\nOn outcomes, the literature genuinely conflicts and I would rather say so than pick a favorable number. Reported reoperation rates range from 7.4% and 9.6%<\/strong> in a large database study to 15% and 21%<\/strong> in a pediatric series \u2014 and the two disagree about which suture technique fares better. The honest parent-facing version: a second procedure is a realistic possibility and a normal part of strabismus care, not evidence that something went wrong.<\/strong><\/p>\nFor intermittent exotropia, international guidance favors nonsurgical management first<\/strong> \u2014 part-time patching, optical correction, monitoring every 6 to 12 months \u2014 with surgery considered for poor control or deteriorating stereoacuity.<\/p>\nDigital and binocular therapies \u2014 real, but narrower than the marketing<\/h3>\nTwo FDA-authorized home systems now exist. Both work on the same principle: show the child engaging video content while handicapping the stronger eye, so both eyes must cooperate.<\/p>\n Luminopia One<\/strong> is a VR headset that streams the child’s own shows, reducing contrast to the stronger eye and applying rotating dichoptic masks. It received a De Novo grant in October 2021<\/strong> for ages 4\u20137 \u2014 the first FDA-authorized digital therapeutic for a neuro-visual disorder \u2014 and a 510(k) clearance in April 2025<\/strong> extended it to ages 4 to under 13 for amblyopia associated with anisometropia and\/or mild strabismus. The pivotal randomized trial (117 children aged 4\u20137, Ophthalmology<\/em> 2022) found significantly greater acuity improvement with the headset plus full-time glasses than with glasses alone at 12 weeks. Note that the age extension rested on real-world registry evidence, not a second randomized trial<\/strong>.<\/p>\nCureSight<\/strong> uses eye tracking and red-blue glasses with a tablet, blurring the fovea of the stronger eye in real time. Its trial (103 children aged 4 to under 9) compared 90 minutes a day, 5 days a week for 16 weeks against 2 hours a day of patching. The result was noninferiority<\/strong> \u2014 comparable improvement with roughly half the treatment hours. That is a meaningful practical finding, and it is worth being precise that noninferior is not the same as better.<\/p>\nAAO’s own bottom line, which the company press releases do not lead with: there is no FDA-approved system that has been proved superior to patching or atropine.<\/strong> PEDIG’s ATS-24 study is designed to make that head-to-head comparison. These devices also require glasses regardless, and are indicated for specific subtypes \u2014 not large-angle strabismus, not deprivation.<\/p>\nOne thing that is not supported<\/h3>\nA joint statement of the AAP, AAO, AAPOS, and the American Association of Certified Orthoptists concludes that scientific evidence does not support eye exercises, behavioral vision therapy, or tinted filters and lenses for improving long-term educational performance<\/strong> in dyslexia and learning disabilities. Vision problems can interfere with learning; they are not the cause of primary dyslexia.<\/p>\nThe distinction is sharp and worth holding onto. Occlusion therapy prescribed by a pediatric ophthalmologist for amblyopia or intermittent exotropia is evidence-based.<\/strong> Open-ended commercial “vision therapy” packages sold for reading and attention problems are not \u2014 and they are marketed hard to exactly the parents reading articles like this one.<\/p>\nFinishing patching is not the finish line<\/h3>\nA PEDIG observational study of successfully treated children under 8 found recurrence in 24% within a year<\/strong> of stopping \u2014 25% after patching, 21% after atropine, most often in the first 13 weeks.<\/p>\nTapering changes that substantially. Among children on 6 to 8 hours a day of patching, recurrence was 42% when treatment stopped abruptly versus 14% when it was first reduced to 2 hours a day.<\/strong> When maximum acuity is reached, the guidance is to taper rather than stop, and to keep monitoring \u2014 beyond two years where risk factors like increasing anisometropia are present.<\/p>\n![\"A](\"https:\/\/wewellinfo.com\/us\/wp-content\/uploads\/2026\/08\/body-8-2.jpg\") <\/p>\n
Photo: Pavel Danilyuk \/ Pexels<\/em><\/p>\n
\nWhat this article is not about: myopia<\/h2>\nA quick boundary, because the two get blurred and the blurring is genuinely unsafe.<\/p>\n Amblyopia and strabismus are neurodevelopmental<\/strong> problems of the visual pathway in early childhood. Myopia progression is a | | | | | | | | | | | | | | |