Capturing myopia and hypermetropia ‘phenotypes’ without formal refraction

Abstract

Understanding genetic and environmental factors that together contribute to the development of myopia is an international research priority. We have investigated the feasibility and accuracy of identifying and classifying refractive error, without formal refraction, as a means of easily identifying affected individuals in a large-scale, non-ophthalmological, and population-based survey. At age 44/45 years, members of the 1958 British birth cohort underwent a biomedical, community survey. Refractive error (autorefraction) was measured and categorised by spherical equivalent (SE) measurement; myopia (SE of -1.00 or more extreme), hypermetropia (+1.00 or more extreme), or emmetropia (-0.99 to +0.99). Lenses of prescribed distance glasses, if worn, were assessed as minifying, magnifying, or making no difference to a standard viewed image and cohort members reported on 'short' or 'long' sightedness. A total of 2499 cohort members, randomly selected, had formal refraction (autorefraction) and 917 (36.7%) of these individuals had their prescribed distance glasses examined. Sensitivities for myopia and hypermetropia using examination of glasses were over 80% and positive predictive values were 95 and 65% respectively whereas self-report of 'short-sightedness' or 'long-sightedness' had poor accuracy. We suggest examination of prescribed distance glasses can be an effective method of 'screening' for refractive error in the field, especially where prevalence is high.