Purpose: To investigate factors associated with spectacle wear in a group of primarily Native-American children provided spectacles free of charge through a school-based vision program.
Methods: Spectacle wear was studied in 247 participants provided two pairs of spectacles the previous year. Univariate and multivariate logistic regression models assessed whether gender, race, parental education levels, family income, uncorrected distance visual acuity, refractive error, or the children's attitudes and beliefs about their vision and spectacles were associated with spectacle wear.
Results: Two thirds of the participants (165/247) were not wearing their spectacles at their annual examination. The most common reasons given for non-wear were lost (44.9%) or broken (35.3%) spectacles. A 1 diopter increase in myopic spherical equivalent was associated with more than a twofold increase in the odds of wearing spectacles [odds ratio (OR) = 2.5, 95% confidence interval (CI) = 1.7 to 3.7]. Among non-myopic participants, increasing amounts of astigmatism in the better- and worse-seeing eye were associated with an increased likelihood of spectacle wear (p ≤ 0.02). In multivariate analysis, only poorer uncorrected acuity in the better-seeing eye (p < 0.001) and shorter acceptance time (p = 0.007) were found to be significantly associated with spectacle wear. For each line of poorer uncorrected acuity in the better-seeing eye, the likelihood that the participant was wearing spectacles increased by 60% (adjusted odds ratio = 1.6; 95% CI = 1.4 to 1.8). Not surprisingly, participants who reported never getting used to their spectacles were less likely to be wearing spectacles than those who reported getting used to wearing glasses in a few days (adjusted OR = 5.7, 95% CI = 1.9 to 17.5).
Conclusions: Despite being provided with two pairs of spectacles, loss and breakage were the most commonly reported reasons for not wearing spectacles. The best predictive factor for determining whether participants were wearing spectacles was their uncorrected acuity.