The AAPOS Position on Vision Therapy: A Fair Look

Parents arrive at our offices having heard some version of the same sentence: vision therapy does not work, or it only helps one narrow diagnosis. Usually the source is a pediatrician or an ophthalmologist relaying the position of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS). That is a fair thing to raise, and it deserves a straight answer rather than a defensive one.

Here is what the AAPOS position actually says, what the research behind it shows, where we think it is right, and where we think it is too blunt for a family trying to sort out a real problem.

Key takeaways

• AAPOS states that behavioral vision therapy has not been proven by science to work, while also stating that orthoptic exercises can help eye-movement problems such as convergence insufficiency (AAPOS glossary, updated 2023).
• The 2009 joint statement and its 2011 technical report are specifically about learning disabilities and dyslexia, and on that question we agree with them (AAP/AAO/AAPOS/AACO, 2009; Handler & Fierson, 2011).
• The evidence is genuinely strong for symptomatic convergence insufficiency in children and for the focusing problems that travel with it (CITT, 2008; Scheiman et al., 2020).
• The evidence is genuinely negative for reading and attention outcomes (CITT-ART, 2019; CITT-ART, 2021).
• The useful question is not whether vision therapy works in the abstract. It is which diagnosis is present and what the evidence shows for that diagnosis.

What AAPOS actually says

The AAPOS patient glossary defines vision therapy as an attempt to develop or improve visual skills and abilities, improve visual comfort and ease, and change visual processing. It then names three main types: behavioral or perceptual vision therapy, therapy aimed at myopia, and orthoptic therapy (AAPOS glossary, updated 2023).

Its verdicts differ by category. Behavioral vision therapy, it says, has not been proven by science to work. Low-power training glasses are described as not very helpful, and there is no evidence that vision therapy slows myopia. Eye exercises have not been shown to improve how children with learning disabilities do in school. But orthoptic exercises, the same page says, can be helpful for eye-movement problems that cause blurry vision, double vision or headaches, naming convergence insufficiency directly (AAPOS glossary, updated 2023).

Most of those skeptical statements are, in our reading, well supported. We do not offer therapy to slow myopia, and we do not tell families that therapy treats a learning disability.

The documents behind the position

The position most often quoted to parents is older and narrower than the glossary. In 2009 the American Academy of Pediatrics, the American Academy of Ophthalmology, AAPOS and the American Association of Certified Orthoptists published a joint statement holding that dyslexia is a language-based disorder, that vision problems can interfere with learning but are not the cause of dyslexia or learning disabilities, and that scientific evidence does not support eye exercises, behavioral vision therapy or tinted lenses as treatments for learning disabilities (AAP/AAO/AAPOS/AACO, 2009).

The 2011 technical report behind it describes dyslexia as arising from deficits in phonologic processing and names evidence-based reading instruction, decoding, fluency, vocabulary and comprehension, as the effective treatment (Handler & Fierson, 2011). The same joint-statement text is carried as an AAO clinical statement, reaffirmed in 2014 (AAO clinical statement, reaffirmed 2014).

Optometry’s organizations hold a different position: that vision problems can and often do interfere with learning, and that optometric care aims to improve visual function as one part of a multidisciplinary approach (American Academy of Optometry & AOA, 1997). The two are not quite answering the same question. One is about the cause of dyslexia; the other is about interference with learning. Families are usually handed only the first.

Where the evidence does support therapy

For symptomatic convergence insufficiency in children, the evidence is not thin. In the 221-child Convergence Insufficiency Treatment Trial, 73 percent of children treated with office-based vergence and accommodative therapy plus home reinforcement were successful or improved after 12 weeks, compared with 35 percent on office placebo, 43 percent with pencil push-ups and 33 percent with home computer therapy (CITT, 2008). A Cochrane network meta-analysis pooling 12 trials and 1,289 participants confirmed the advantage over placebo in children, with a risk ratio of 3.04 (Scheiman et al., 2020). The adult evidence in that same review is thinner and less clear.

The focusing side holds up too. Roughly three-quarters of the CITT children also had accommodative dysfunction, and accommodative amplitude improved 9.9 diopters with office-based therapy versus 2.2 with placebo (Scheiman et al., 2011). A later CITT-ART analysis found amplitude improving 8.6 versus 5.2 diopters, with 69 percent versus 32 percent of children reaching a normal amplitude (Chen et al., 2021). This is the part of the glossary we would ask parents to read twice, because AAPOS itself acknowledges it.

Where the evidence does not support it

Here is the part our own profession sometimes skips. CITT-ART enrolled 310 children with symptomatic convergence insufficiency and tested the reading question directly. Reading comprehension improved 3.7 points with 16 weeks of office-based therapy and 3.8 points with placebo therapy. Therapy was no better than placebo (CITT-ART, 2019). A companion analysis of the same children found no significant difference between the groups on the parent-reported inattention scale; both groups improved, which the authors attributed to nonspecific effects rather than to the vergence and accommodative gains (CITT-ART, 2021).

The treatment paper from the same trial belongs alongside them. Clinical signs improved substantially more with therapy, with near point of convergence gaining 10.4 versus 6.2 centimeters and positive fusional vergence 23.2 versus 8.8 prism diopters, but symptom scores improved about equally in both groups (CITT-ART, 2019). Read the 2008 and 2019 trials together and the honest summary is this: office-based therapy reliably changes the measurements, changed symptoms in the first trial but not the second, and does not change reading scores. We say all of that out loud. More detail is in our walkthrough of what the CITT and CITT-ART trials actually found.

An independent review reaches a similarly split verdict. Evaluating behavioral optometry across ten treatment areas, Barrett found the evidence consistent with practitioners’ claims for convergence insufficiency, yoked prisms in neurological patients and vision rehabilitation after brain injury, but concluded that a large majority of behavioral approaches were not evidence-based and could not be advocated (Barrett, 2009). That review is not a defense of our field. It is a reasonable map of it.

The dyslexia point is right, and often misheard

Dyslexia is not caused by an eye problem, and vision therapy is not a treatment for it. We agree with AAPOS, the AAP and the AAO on that, and the population data support them: in a cohort of 5,822 children, four of every five with severe reading impairment had entirely normal ophthalmic function, and no association was found with convergence or accommodation (Creavin et al., 2015).

What families often hear, though, is a broader claim: that vision has nothing to do with reading difficulty at all. That is not what the statements say, and the clinic-based data are less tidy. Comparing 29 children with developmental dyslexia to 33 typical readers, one study found some visual deficit in 79 percent versus 33 percent, while explicitly noting that the cause and clinical relevance were uncertain (Raghuram et al., 2018). And when children with learning disorders and binocular anomalies received vision therapy, what improved was the binocular measurements, not reading (Hussaindeen et al., 2018).

So a child can have a language-based reading disorder and an uncomfortable focusing or convergence problem. Treating the second does not treat the first. It may make sustained near work less punishing while the reading instruction that actually works does its job. That is the whole claim, and it is deliberately a modest one.

Function versus pathology: our clinical perspective

This next part is our view at Vision & Learning Center, not a research finding, and we want the difference to be obvious.

AAPOS speaks from pediatric ophthalmology and strabismus, a discipline built around ocular disease, alignment, amblyopia and surgical care. That lens is legitimate, and its guidance is strongest where pathology and alignment are the question. Our lens asks something adjacent: with healthy eyes and clear acuity, how comfortably does the visual system hold up through a school day?

A child can have 20/20 sight, a normal eye exam, and still lose their place, get headaches after twenty minutes, or see print swim at the end of a page. Some of those complaints have real treatment evidence behind them. Others do not. Our obligation is to say which is which at the evaluation, before anyone commits to a program.

A more precise message for parents

If we could rewrite the message families receive, it would read something like this:

• Vision therapy does not treat dyslexia or learning disabilities (AAP/AAO/AAPOS/AACO, 2009; Handler & Fierson, 2011), attention problems (CITT-ART, 2021) or myopia progression (AAPOS glossary, updated 2023).
• Vision therapy does have randomized, placebo-controlled support for symptomatic convergence insufficiency (CITT, 2008; Scheiman et al., 2020) and the accommodative dysfunction that often accompanies it (Scheiman et al., 2011; Chen et al., 2021).
• Vision therapy did not beat placebo for reading comprehension (CITT-ART, 2019) or parent-reported inattention (CITT-ART, 2021) in the largest trial that tested it.
• Everything else sits on smaller studies, case series or clinical reasoning, and should be described that way when it is offered.
• A functional visual complaint deserves to be measured before it is dismissed, and measured again before therapy is continued.

That message protects families from overselling and from premature dismissal at the same time. Our longer take is in does vision therapy work: an honest evidence summary and in six vision therapy myths checked against evidence. If you are still working out what therapy involves, start with what is vision therapy, and for the wider picture see our guide to vision and learning.

Where to go from here

AAPOS is right to demand evidence and right that dyslexia is not an eye problem. Where we think the public message falls short is in how easily a parent can read it and conclude that any visual complaint outside a surgical or disease model is not worth measuring.

So bring the skepticism with you. Ask what the diagnosis is, what the evidence shows for it, what will be measured, how often, and what happens if the numbers do not move. Those questions will tell you more about a practice than any position statement will.

If your child is losing their place while reading, getting headaches with homework, seeing double at near, or avoiding close work in a way that does not look like ordinary reluctance, a comprehensive developmental vision evaluation will tell you whether a treatable visual problem is part of the picture, and whether it is not. Schedule an evaluation or see what our examinations include.