What the CITT and CITT-ART Trials Actually Found

Parents ask a fair question when we recommend vision therapy: is there real evidence for this, or is it something optometrists simply believe in? For one condition — convergence insufficiency — the answer is unusually clear, because the treatment has been put through two large multicentre randomised trials with genuine placebo arms.

Those trials are the Convergence Insufficiency Treatment Trial (CITT, 2008) and the Convergence Insufficiency Treatment Trial – Attention and Reading Trial (CITT-ART, 2019 and 2021). Both are quoted often, usually by people who have read only one of them. Here is what each actually found, including the parts that do not flatter our profession.

Key takeaways

• In CITT, 73% of children were successful or improved after 12 weeks of office-based vergence and accommodative therapy, versus 35% with placebo therapy (CITT Study Group, 2008).
• Among children who became asymptomatic, 84.4% of the office-therapy group were still asymptomatic a year later (CITT Study Group, 2009).
• CITT-ART confirmed that the clinical signs improve with therapy, but symptom scores improved about the same amount in the placebo group (CITT-ART, 2019).
• CITT-ART found no benefit over placebo for reading comprehension (CITT-ART, 2019) or for attention (CITT-ART, 2021).
• A Cochrane network meta-analysis still concludes office-based therapy beats placebo in children (Scheiman et al., 2020).
• The honest summary: the evidence supports treating symptomatic convergence insufficiency. It does not support promising better reading, grades or attention.

Why these two trials matter

Convergence insufficiency is difficulty sustaining the inward eye turn that near work demands. It is common — about 13% of fifth- and sixth-graders in a US school-based study met a high-suspect or definite definition (Rouse et al., 1999) — and its symptoms overlap heavily with other things, which is exactly why good trials were needed. For more on how often it occurs, see how common is convergence insufficiency.

Testing a behavioural treatment fairly is hard, because children improve with time, attention and expectation regardless of what you do. Both trials handled that the same way: a placebo therapy program delivered in the office on the same schedule, with activities designed not to build convergence or focusing. Between them the two trials randomised more than 500 children. For scale, the Cochrane review of this entire field pooled 12 trials and 1289 participants in total (Scheiman et al., 2020).

Symptoms were measured with the Convergence Insufficiency Symptom Survey, a questionnaire validated in children aged 9 to 18, where a score of 16 or higher separated children with convergence insufficiency from those without with 96% sensitivity and 88% specificity (Borsting et al., 2003).

CITT (2008): the trial that changed the conversation

A pilot study in 47 children set the design (Scheiman et al., 2005). The full trial then randomised 221 children aged 9 to 17 with symptomatic convergence insufficiency to one of four 12-week programs: office-based vergence and accommodative therapy with home reinforcement, home pencil push-ups, home computer vergence therapy plus push-ups, or office-based placebo therapy.

After 12 weeks, mean symptom scores were 15.1 for office-based therapy, 24.7 for pencil push-ups, 21.3 for home computer therapy and 21.9 for placebo. Rates of a successful or improved outcome were 73%, 43%, 33% and 35% respectively (CITT Study Group, 2008).

How to read that result honestly

Three things deserve to be said in the same breath.

1. The advantage over placebo was real and large — roughly 38 percentage points. This is not a marginal finding.
2. About a third of the placebo group also improved. Structured attention, weekly appointments and time do a lot on their own. One child getting better is never proof that therapy caused it.
3. The primary outcome was a symptom questionnaire, not a reading test, a grade or a behaviour rating.

That third point matters more than it first appears, because the symptoms children report are not purely ocular. In the CITT sample, performance-related complaints such as losing place and trouble concentrating were actually rated higher at baseline than eye-related complaints like blur or double vision (Barnhardt et al., 2012). A questionnaire that improves is telling you a child is more comfortable at near — not that their academic performance changed.

One year later: did the gains hold?

Seventy-nine children who had become asymptomatic were followed for a year. In the office-based therapy group, 84.4% were still asymptomatic at 12 months and 87.5% were still classified as successful or improved, with no significant drift in symptom scores during follow-up (CITT Study Group, 2009).

The caveat is important: only children who had already responded were followed, and the comparison groups were small. This tells us that gains tend to hold in responders. It does not tell us what proportion of all treated children relapse.

CITT-ART (2019): the harder question

CITT-ART asked whether treating convergence insufficiency changes anything beyond eye comfort. It enrolled 311 children aged 9 to 14 and randomised them to 16 weeks of office-based vergence and accommodative therapy or placebo therapy (CITT-ART, 2019).

The clinical signs improved clearly

Near point of convergence improved by 10.4 cm with therapy versus 6.2 cm with placebo, and positive fusional vergence improved by 23.2 versus 8.8 prism dioptres, both differences highly significant (CITT-ART, 2019). Whatever else the trial showed, therapy measurably changed how well these children’s eyes converged and held convergence.

The symptom scores did not separate from placebo

In the same paper, symptom scores improved by 11.8 points with therapy and 10.4 points with placebo, a difference that was not statistically significant (CITT-ART, 2019). Both groups felt better; the therapy group did not feel meaningfully better than the placebo group.

That is a different result from 2008, and it belongs beside the 2008 numbers every time they are quoted. The two trials also differed in ways that may matter. Both required the same entry symptom score of 16 or higher, but CITT-ART narrowed the age range to 9 to 14 in grades 3 to 8, excluded children scoring below 70 on a nonverbal reasoning measure or below 80 on a word-reading measure, and ran 16 weeks rather than 12 (CITT-ART, 2019; CITT Study Group, 2008). But the plain fact is that the second large trial did not reproduce the symptom advantage of the first.

Reading comprehension: no benefit

In 310 of those children, reading comprehension on a standardised test improved by 3.7 points with therapy and 3.8 points with placebo (CITT-ART, 2019). That was a large randomised trial, with a placebo arm, built to answer exactly the question parents ask — and the answer was no. We will not tell you that vision therapy raises reading scores, because the evidence says it does not.

Attention: no benefit

The attention paper reported no significant difference between groups on parent-rated inattention (effect size 0.036, confidence interval −0.21 to 0.28) or hyperactivity and impulsivity (−0.003), with similar results for teacher reports, a homework problems checklist and a sustained-attention test (CITT-ART, 2021).

The detail worth sitting with is what else that paper found: both groups showed large improvements in inattention, homework completion and sustained attention. Parents in the placebo arm saw their children concentrate better and fight less about homework too. The authors attribute those gains to the nonspecific effects of an intensive therapy regimen rather than to changes in convergence — a conclusion any honest practitioner has to take seriously.

Focusing, prism and adults

Convergence rarely travels alone. Of the 221 CITT children, 164 (74%) also had an accommodative (focusing) problem. Accommodative amplitude increased by 9.9 dioptres with office-based therapy versus 2.2 with placebo, and reduced amplitude had recurred in only 12.5% of children a year later (Scheiman et al., 2011). For focusing problems occurring on their own, the picture is much thinner: a review found prevalence estimates ranging from under 1% to 61.6% depending on criteria and no high-quality evidence for a standard treatment protocol (Hussaindeen & Murali, 2020).

Prism is the other treatment families ask about, and the evidence is age-dependent. Base-in prism reading glasses were no better than placebo glasses in 72 children (Scheiman et al., 2005), while base-in prism progressives did outperform placebo progressives in 29 symptomatic adults aged 45 to 68 (Teitelbaum et al., 2009). We work through that trade-off in prism glasses vs vision therapy.

Adults sit in the least certain part of the evidence. In a trial of 46 adults aged 19 to 30, only office-based therapy significantly changed the clinical signs, but symptom elimination was reached by 42% on therapy versus 31% on placebo, and more than half of treated patients remained symptomatic (Scheiman et al., 2005). The Cochrane review reaches the same conclusion: strong evidence in children, unclear in adults (Scheiman et al., 2020).

What this means if you are deciding about therapy

Statements the evidence supports:

• Office-based vergence and accommodative therapy with home reinforcement is the best-supported treatment for symptomatic convergence insufficiency in children.
• It reliably improves the measurable signs — near point of convergence and fusional vergence — well beyond placebo.
• In children who respond, gains generally hold for at least a year.
• It also improves focusing amplitude in children who have both problems.

Statements the evidence does not support, and that we will not make:

• That treating convergence insufficiency improves reading comprehension or grades.
• That it improves attention or treats ADHD.
• That any given child is certain to succeed.

Our clinical perspective at Vision & Learning Center. We recommend therapy when a child is genuinely symptomatic and the measurements support the diagnosis, and we define success as comfort and endurance at near work — being able to get through a full class period of reading without the words swimming, the head aching or the book closing. When a child also struggles to read, we say plainly that vision therapy is not reading instruction, and we work alongside teachers and reading specialists rather than in place of them. Because the placebo arms in both trials improved, we track objective signs at each progress check, not just how a child says they feel.

If reading, homework or screen work has become uncomfortable for your child or for you, the place to start is a comprehensive developmental vision evaluation that measures alignment, convergence, focusing and eye movements together, and a conversation about what treatment can and cannot be expected to do. Schedule an evaluation, see what our examinations include, or read our full guide to convergence insufficiency: symptoms, diagnosis and treatment.