PEDIG 2016.

Study characteristics
Methods	Study design: RCT Number randomized: 204 (75 assigned to home‐based computer vergence; 85 assigned to home‐based near target push‐up; 44 assigned to home‐based placebo) Unit of randomization: individual participant (convergence insufficiency is a binocular vision disorder) Number analyzed: 169 (83%) (69 of 75 assigned to home‐based computer vergence; 69 of 85 assigned to home‐based near target push‐up; 31 of 44 assigned to home‐based placebo) Number of centers: 30 Date of first enrolment: June 2012 Length of follow‐up: planned: 12 weeks after initiation of treatment; actual: 12 weeks after initiation of treatment Sample size estimation: "The pre‐planned sample size was 595 participants (238 participants in each of the two active treatment groups, 119 participants in the placebo (HB‐P) group) to have 90% power to detect a treatment group difference for each of the two pairwise comparisons, HB‐C versus HB‐PU and HB‐C versus placebo, assuming true population success percentages of 30%, 15% and 10% for the HB‐C group, HB‐PU group, and HB‐P group, respectively, with a type I error rate of 2.5% per comparison (5% overall) including adjustments for 3 planned interim analyses for futility and no more than 10% loss to follow‐up. The assumed success percentages were determined based on the Convergence Insufficiency Treatment Trial (CITT) and clinical expertise. Due to sample size considerations, it was not feasible to compare the HB‐PU group with the HB‐P group as a primary outcome pairwise comparison based on the assumed successful outcome percentages of 15% versus 10%."
Participants	Country of recruitment: United States Mean age: 12.2 ± 2.4 (SD) years in home‐based computer vergence group; 12.6 ± 2.5 (SD) years in home‐based near target push‐up group; 12.3 ± 2.3 (SD) years in home‐based placebo group Sex: 61% were female in home‐based computer vergence group; 54% were female in home‐based near target push‐up group; 59% were female in home‐based placebo group Key inclusion criteria: age 9 to <18 years; near exophoria 4 Δ or greater than at distance; reduced positive fusional vergence at near (PFV), defined as < 20 Δ mean PFV blur or failing Sheards’ criterion (mean PFV measured less than twice the near phoria magnitude); mean near point of convergence (NPC) of ≥ 6 cm break; symptomatic convergence insufficiency, defined as a Convergence Insufficiency Symptom Survey (CISS) score of ≥ 16 points; best‐corrected visual acuity of 20/25 or better in each eye at distance and near; near random dot stereoacuity of at least 400 seconds of arc Key exclusion criteria: ≥ 2 logMAR line difference in best‐corrected visual acuity between the two eyes; constant or intermittent exotropia at distance; constant exotropia at near; any esotropia at distance or near; distance exophoria > 10 Δ; history of strabismus surgery; anisometropia ≥ 2.00 D in any meridian between the eyes; prior intraocular or refractive surgery; primary vertical heterophoria greater than 1 Δ; diseases known to affect accommodation, vergence, and ocular motility such as multiple sclerosis, Graves orbitopathy, myasthenia gravis, diabetes mellitus, or Parkinson disease; current use of any ocular or systemic medication known to affect accommodation or vergence such as anti‐anxiety agents; near point of accommodation > 20 cm in the right eye; manifest or latent nystagmus evident clinically; history of chronic headaches unrelated to reading activity
Interventions	Intervention regimen #1: home‐based computer vergence/accommodative therapy (HB‐C) The HB‐C group was prescribed 15 minutes of active computer vergence/accommodative therapy (CVAT) and 5 minutes of placebo flipper exercises. Intervention regimen #2: home‐based near target push‐up therapy (HB‐PU) The HB‐PU group was prescribed 15 minutes (in full or split into three 5‐minute intervals) of a well‐defined near target push‐up (NTP) procedure and 5 minutes of placebo CVAT. Intervention regimen #3: home‐based placebo treatment (HB‐P) The HB‐P group was prescribed 15 minutes of placebo CVAT and 5 minutes of placebo flipper exercises.
Outcomes	Primary outcome: successful outcome if meets at 12 weeks success criteria for all of the followings: (1) Convergence Insufficiency Symptom Survey score of < 16 points and improvement of ≥ 9 points since baseline; (2) mean near point of convergence (NPC) break of < 6 cm and a 12‐week to baseline ratio of < 0.763 for mean NPC break; (3) mean positive fusional vergence at near blur of >15 Δ and a 12‐week to baseline ratio of > 1.419 for mean PFV blur. Key secondary outcomes: percentage of participants who met success criteria of the original components of the primary outcome for both vergence measures (NPC and PFV). and percentage classified as "improvers" (12‐week CISS score improvement of ≥9 points since baseline; 12‐week to baseline ratio of > 0.763 for mean NPC break, and 12‐week to baseline ratio of > 1.419 for mean PFV blur. No harms were reported.
Notes	Funding sources: National Eye Institute, National Institutes of Health, Bethesda, Maryland, USA. Subgroup analyses: none reported Trial registration: NCT01515943
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Participants were randomly assigned (using a permutated block design stratified by site) to 1 of 3 HB treatment groups (HB‐C group, HB‐PU group, and HB‐P group) in a 2:2:1 ratio with a 1 in 5 chance of being randomized to the HB‐P group"
Allocation concealment (selection bias)	Low risk	Allocation was adequately concealed (personal communication with the investigator).
Blinding (performance bias and detection bias) Primary outcome	Low risk	"Participants were to remain masked to their treatment group until they completed the study." "Placebo CVAT was similar to the active version except there was no specific accommodation program and the procedures were designed not to stimulate or exert any extra demand on the vergence system."
Blinding (performance bias and detection bias) Secondary outcomes	Low risk	"With the exception of the cycloplegic refraction and stereoacuity testing, the CISS and clinical testing were repeated at each follow‐up examination by an examiner who was masked to participants’ treatment group". "Examiners were masked to treatment group for all examinations."
Incomplete outcome data (attrition bias) Primary outcome	High risk	Differential loss to follow‐up [6(8%) participants in home‐based computer vergence group, 16(19%) in home‐based near target push‐up, and 13(30%) in home‐based placebo therapy] who were randomized were not included in the final analysis.
Incomplete outcome data (attrition bias) Secondary outcomes	High risk	See above.
Selective reporting (reporting bias)	Low risk	All prespecified outcomes were reported