Evaluating masking in a randomized, double-masked clinical trial in children with myopia

The purposes of this study were to evaluate the success of masking in the Correction of Myopia Evaluation Trial (COMET), a study designed as a randomized, double-masked clinical trial comparing the effect of progressive addition lenses (PALs) with single-vision lenses (SVLs) in slowing myopia progression among children and to investigate the impact of unmasking on the treatment effect.

The success of masking of children and parents was assessed by questionnaires administered after the clinical trial results were reported and before families were informed of their child's lens assignment. Children and parents who correctly guessed their assignment with absolute or "pretty sure" certainty were considered as unmasked. The role of children's age, gender, and ethnicity in unmasking was evaluated using logistic regression. The impact of unmasking on the treatment effect at 3 years was investigated using analysis of variance. The success of masking of the study optometrists was assessed by data collected at the child's follow-up visits.

Overall 93% (436 of 469) of the COMET children responded to the questions about lens assignment and certainty. Thirty-seven percent (163 of 436) of children met the criteria for being unmasked with similar results from the parents. The majority (91%) of the 163 "unmasked" children reported that their glasses were the main reason for lens identification. Although children 10 to 11 years at baseline became unmasked more often than did children 6 to 9 years (44% vs. 31%, p = 0.01), there was no difference between boys and girls becoming unmasked (p = 0.2). The treatment effect did not differ between the masked and unmasked children (p = 0.69). The study optometrists became unmasked for five children over the course of 3-year follow up.

Masking was successfully maintained for the study optometrists and the majority (63%) of COMET children. Although some children were able to identify their lens assignment, the primary study outcome was not affected by unmasking.