JMIR Publications

ABSTRACT

Background:

Patching is a standard treatment for childhood amblyopia but is frequently limited by adherence challenges. Home-based binocular digital therapies offer an alternative treatment format and can enable automated monitoring of treatment use, although randomized evidence directly comparing active gamified interventions with patching in children younger than 7 years remains limited. Whether baseline amblyopic-eye visual acuity modifies comparative treatment response is also uncertain.

Objective:

To determine whether home-based gamified binocular therapy is noninferior to part-time patching for improving amblyopic-eye best-corrected visual acuity (BCVA) in children aged 4 to <7 years with unilateral amblyopia, and to explore treatment-effect heterogeneity according to baseline amblyopic-eye acuity.

Methods:

Prospective, multicenter, randomized, evaluator-masked, parallel-group noninferiority trial conducted at 3 tertiary centers in Shanghai, China, enrolling 144 children with unilateral amblyopia (amblyopic-eye BCVA 20/32 to 20/200).Participants were randomized 1:1 to 8 weeks of home-based gamified binocular therapy (1 hour/day, 5 days/week) or part-time patching (2 hours/day, 7 days/week). Treatment use in the binocular therapy group was recorded automatically by the system, whereas patching duration was recorded by caregivers in paper diaries.The primary outcome was mean change in amblyopic-eye BCVA from baseline to week 8, analyzed using a mixed model for repeated measures, with a noninferiority margin of 0.10 logMAR. Secondary outcomes included stereoacuity and recorded treatment adherence.

Results:

Among 136 participants who received assigned treatment (mean age, 4.76 years; 49.3% female), mean BCVA improvement at week 8 was 0.15 logMAR in the binocular therapy group and 0.13 logMAR in the patching group. The adjusted least-squares mean difference was 0.01 logMAR (95% CI, -0.02 to 0.04), establishing noninferiority (lower bound > -0.10). A significant treatment-by-baseline-BCVA interaction was observed (P = 0.013), and Johnson-Neyman analysis identified a data-derived threshold near 0.48 logMAR. Consistent with this finding, among 37 participants with baseline BCVA ≥0.50 logMAR, mean improvement was greater with binocular therapy than with patching (0.25 vs. 0.15 logMAR; P = 0.017), and more participants achieved a ≥2-line gain (89.5% vs. 50.0%; P = 0.013). Recorded adherence was 99.8% in the binocular therapy group based on device-recorded treatment minutes and 87.6% in the patching group based on caregiver diaries (P = 0.007). Stereoacuity outcomes did not consistently favor binocular therapy.

Conclusions:

Home-based gamified binocular therapy was noninferior to part-time patching for improving amblyopic-eye BCVA at 8 weeks and enabled automated recording of treatment use. Exploratory analyses suggested that relative treatment effects varied according to baseline acuity, but the data-derived threshold and subgroup findings require prospective validation before informing treatment selection. Clinical Trial: Chinese Clinical Trial Registry; ChiCTR2300079090