Flitcroft’s model of refractive development in childhood and the possible identification of children at risk of developing significant myopia

Abstract

PurposeTo better understand juvenile myopia in the context of overall refractive development during childhood and to suggest more informative ways of analysing relevant data, particularly in relation to early identification of those children who are likely to become markedly myopic and would therefore benefit from myopia control. MethodsExamples of the frequency distributions of childhood mean spherical refractive errors (MSEs) at different ages, taken from previously-published longitudinal and cross-sectional studies, are analysed in terms of Flitcroft’s model of a linear combination of two Gaussian distributions with different means and standard deviations. Flitcroft hypothesises that one, relatively-narrow, Gaussian (Mode 1) represents a “regulated” population which maintains normal emmetropisation and the other, broader, Gaussian (Mode 2) a “dysregulated” population. ResultsAnalysis confirms that Flitcroft’s model successfully describes the major features of the frequency distribution of MSEs in randomly-selected populations of children of the same age. The narrow “regulated” Gaussian typically changes only slightly between the ages of about 6 and 15, whereas the mean of the broader “dysregulated” Gaussian changes with age more rapidly in the myopic direction and its standard deviation increases. These effects vary with the ethnicity, environment and other characteristics of the population involved. At all ages there is considerable overlap between the two Gaussians. This limits the utility of simple refractive cut-off values to identify those children likely to show marked myopic progression. ConclusionsAnalysing the frequency distributions for individual MSEs in terms of bi-Gaussian models can provide useful insights into childhood refractive change. A wider exploration of the methodology and its extension to include individual progression rates is warranted, using a range of populations of children exposed to different ethnic, environmental and other factors.