Toward early intervention based on age-specific vision checkups: A vision impairment survey in Yantai, China.

Abstract

The prevalence of myopia among children and adolescents is currently rising to alarming levels (>80%) in China. This study used several routinely collected demographic factors to quantify myopia and glass-wearing rates for primary and secondary school students. We identified myopia risk factors and proposed new aspects for early intervention. This study was a cross-sectional survey of myopia and glass-wearing rates for students (6-18 years old) in Yantai, China. We collected both vision (vision acuity [VA] and spherical equivalence [SE]) and glass-wearing information to establish respective logistic models for quantifying myopia and glass-wearing rate. We further propose a joint decision region (VA, SE, age) to guide early intervention. Among 10,276 children, 63% had myopia (65% wore glasses). The prevalence of myopia increases with age and levels off during adulthood. Females had a higher overall prevalence rate than males (P < .001). The rural age mode (≈15.5) is about 2 years larger than the urban age (≈13.5) for myopia students. For the myopia rate, in the age ≤14.5, the linear age effect was significant (odds ratio [OR] = 1.73, P < .0001), males had a significant negative baseline effect at the start of schooling (vs. females) (OR = 0.68, P < .0001), and the urban group had a significant positive baseline effect (vs. rural) (OR = 1.39, P < .0001). The correlation between VA and SE increases with age and has a directional shift (from negative to positive) at ages 8 to 9. For the glass-wearing rate, age had a significant positive effect (OR = 1.25, P < .0001), VA had a significant negative effect (OR = 0.002, P < .0001), and body mass index had a slightly significant positive effect (OR = 1.02, P = .03). Urban female have a higher myopia rate than rural male at the start of schooling, and vocational high school has improved vision upon high school. Body mass index was not a significant factor for myopia. The myopia rate model is specific to age range (separated at 14.5 years old). Students of lower ages are less likely to wear glasses for correction, and this may require intervention. The temporal age-specific (VA, SE) correlations and joint distributions strengthen the speculation in the literature that age 8 to 9 is a critical intervention period and motivates us to propose a rigorous intervention decision region for (age, VA, and SE) which mainly applies for this tight age period.