Astigmatism and emmetropization in a native American population

Previous research has shown a high prevalence of astigmatism among preschool-age, school-age, and adult members of some Native American tribes (1,2). Data from cross-sectional studies indicate that prevalence of astigmatism decreases or remains stable during the school years in some tribal groups (3-5). However, longitudinal data are needed to determine whether emmetropization of astigmatism occurs. Subjects in the present study were 208 three- to five-year-old children who were participants in the Tohono O'Odham Nation's Head Start program. Each child had cycloplegic refractive error measured with the Nikon Retinomax K-Plus autorefractor on at least three occasions during two academic years. For each child's dataset, an estimate of slope was calculated and compared statistically to a slope of zero (6,7). Results showed that the 69 subjects with high baseline astigmatism (>1.50 D) were more likely than the 139 subjects with low astigmatism (<1.50 D) to have decreasing (odds ratio=1.68, p=0.003) or increasing (odds ratio=1.39, p=0.002) slopes over time. However, most subjects (63.8% of the high astigmats and 87.8% of the low astigmats) showed no change in amount of astigmatism over time. Average change was 0.04 D/year (SD 0.36) in the high astigmatism group, compared to 0.02 D/year (SD 0.18) in the low astigmatism group, which was not a significant difference. Thus, we found little evidence for emmetropization of astigmatism during the preschool years. Follow-up of refractive error development in this population of children during elementary school is in progress, and data will be presented concerning whether the failure of emmetropization seen in the preschool years continues as children get older.