Prevalence and predictors of refractive error in a genetically isolated population: the Norfolk Island Eye Study

Background: We aimed to determine the prevalence and associations of refractive error on Norfolk Island. Design: Population-based study on Norfolk Island, South Pacific. Participants: All permanent residents on Norfolk Island aged $15 years were invited to participate. Methods: Patients underwent non-cycloplegic auto- refraction, slit-lamp biomicroscope examination and biometry assessment. Only phakic eyes were analysed. Main Outcome Measures: Prevalence and mul- tivariate associations of refractive error and myopia. Results: There were 677 people (645 right phakic eyes, 648 left phakic eyes) aged $ 15 years were included in this study. Mean age of participants was 51.1 (standard deviation 15.7; range 15–81). Three hundred and seventy-six people (55.5%) were female. Adjusted to the 2006 Norfolk Island popula- tion, prevalence estimates of refractive error were as follows: myopia (mean spherical equivalent $-1.0 D) 10.1%, hypermetropia (mean spherical equivalent $ 1.0 D) 36.6%, and astigmatism 17.7%. Significant independent predictors of myopia in the multivariate model were lower age (P < 0.001), longer axial length (P < 0.001), shallower anterior chamber depth (P = 0.031) and increased corneal curvature (P < 0.001). Significant independent pre- dictors of refractive error were increasing age (P < 0.001), male gender (P = 0.009), Pitcairn ances- try (P = 0.041), cataract (P < 0.001), longer axial length (P < 0.001) and decreased corneal curvature (P < 0.001). Conclusions: The prevalence of myopia on Norfolk Island is lower than on mainland Australia, and the Norfolk Island population demonstrates ethnic dif- ferences in the prevalence estimates. Given the sig- nificant associations between refractive error and several ocular biometry characteristics, Norfolk Island may be a useful population in which to find the genetic basis of refractive error.