Interventions to increase time spent outdoors for preventing incidence and progression of myopia in children (Cochrane systematic review)

5 RCTs (4 cluster-randomized), 10,733 participants

Primary school children, mostly grades 1-2 (aged 6-9); searches to 24 June 2022

The only review of this literature restricted to randomized designs - every other synthesis pools cohort and cross-sectional studies and therefore inherits their confounding. GRADE certainty assessed per outcome and per timepoint. Prespecified clinical importance thresholds: 0.1 D for refractive change and 3 percentage points for incidence. Graded B rather than A because it pools only five trials, several with cluster-randomization problems, and certainty is low at most timepoints.

Myopia incidence: 7.1% vs 9.5% at 1 year (RR 0.82, 95% CI 0.56-1.19, low certainty); 22.5% vs 26.7% at 2 years (RR 0.84, 95% CI 0.72-0.98, moderate certainty); 30.5% vs 39.8% at 3 years (RR 0.77, 95% CI 0.59-1.01, moderate certainty). Refractive change: MD 0.08 D at 1 year (95% CI -0.01 to 0.17, low certainty), 0.13 D at 2 years (95% CI 0.06 to 0.19, moderate certainty), 0.17 D at 3 years (95% CI -0.17 to 0.51, low certainty). Axial length MD -0.04 mm at 2 years (95% CI -0.07 to -0.01). The pattern is that only the 2-year estimates clear both the certainty and clinical-importance bars; the 1-year and 3-year confidence intervals include no benefit. No included trial reported unaided visual acuity or quality of life, and no adverse events were reported.

Low. Restriction to RCTs means the pooled estimate is not vulnerable to the passive gene-environment correlation that contaminates the observational outdoor-time literature (families who send children outdoors differ heritably from those who do not).