Free Spectacles Change the Starting Odds
The specific question for refractive-error services here is what changes when spectacles are provided free of charge. The randomized evidence applies to children with refractive errors and measures reported or observed spectacle wearing, not every form of visual or educational benefit. Reading the result carefully supports a useful judgment: removing price can improve a downstream outcome, but it does not turn the whole care pathway into a guaranteed success.
What the randomized evidence actually compares
The Frontiers study is a systematic review and meta-analysis of randomized controlled trials. That design matters because randomization can make the groups being compared more similar at the start than they would be in a simple survey of spectacle owners. It does not eliminate every limitation, but it gives the free-versus-comparison question more causal credibility than an observation made after families have already chosen whether to seek or buy spectacles.
Fourteen articles met the review's inclusion criteria, and 11 studies supplied enough comparable information for the meta-analysis. The studies involved children under 18 with refractive errors and evaluated interventions that provided free spectacles. The comparison conditions were not all identical, but the review specifically contrasts free spectacles with providing only a prescription or a prescription plus a letter to parents. The outcome was compliance with spectacle use at the end of the relevant follow-up, measured either by observation or by children's self-report, depending on the study.
That description of the comparison prevents a common overstatement. The trial question was not simply whether spectacles are useful, nor whether every form of eye care should be free. It was whether removing the price of spectacles, within the service arrangements used in those studies, changed the likelihood of the measured wearing outcome relative to less-complete forms of provision. The answer can be important without answering the larger questions about the best assessment method, the best frame, or the effect of spectacles on every aspect of a child's life.
Across the synthesis, 4,124 of 7,765 children were reported as compliant with spectacle use, an overall compliance of 53.11%. That number is easy to misread. It is not the percentage of all children in the world who wear spectacles, nor is it a universal retention rate for every free-spectacle program. It is the pooled result within the included evidence base, using the definitions and follow-up periods chosen by those studies. It tells us that wearing was common enough to measure and important enough to improve, but also that nearly half of the children in that pooled denominator were not counted as compliant at the relevant endpoint.
The review reports an odds ratio of 2.45 for the effect of receiving free spectacles, with a 95% confidence interval from 1.39 to 4.30. In plain language, the odds of the measured compliance outcome were higher in the free-spectacle condition than in the comparison conditions. The confidence interval expresses uncertainty around the estimated contrast: the data are compatible with a smaller increase than the point estimate and with a larger one, within the statistical model used by the review.
An odds ratio is not the same as a percentage-point increase, and it is not the same as saying that 2.45 times as many children wore glasses. To translate an odds ratio into an absolute probability, one needs a baseline probability for the comparison group.