Abstract
Purpose :
There is now substantial evidence to indicate that environmental factors including time outdoors can influence the development of myopia, particularly in school-aged children. It is well known that urban and rural differences in myopia prevalence exist, but few studies have attempted to characterise the built or natural environment using objective means. The aim of this study is to investigate the relationship between the prevalence of myopia and the quantity of green spaces across different regions of the world, using an objective satellite imaging technique.
Methods :
The prevalence of myopia in the 15 to 19 year age group in Australia, Brazil, China, Finland, India, Iran, Japan, Oman, Singapore, South Africa, and the United Kingdom were collected from study data obtained from a systematic review and meta-analysis by Holden et al. (2016). Normalised Difference Vegetation Index (NDVI), an indicator of vegetation density, was derived from Landsat 7 Enhanced Thematic Mapper Plus (ETM+) satellite data. Green space (mean NDVI) was quantified using a 30-kilometre radius buffer surrounding point coordinates corresponding to locations mentioned in the Holden et al. prevalence studies (n = 12) that contributed most to prevalence data. Simple linear regression was used to analyse yearly data, whilst a generalised linear model (GLM) was fit to analyse aggregated NDVI and prevalence data. A mixed effects model was applied to assess the significance of green space when study was a random effect.
Results :
The results indicated a negative cubic relationship between surrounding green spaces and myopia prevalence, although the association was weak. The results of the mixed effects model suggested that green space was not significant when the study effects were considered as a random factor (p = .099).
Conclusions :
There was no statistically significant association between green space and myopia prevalence in the 15-19-year-old group. Use of a larger sample and standardisation of green space data may improve the robustness of the data. Additionally, use of buffer zones with smaller radii should be considered as they may be more applicable to real life daily green space exposure and potentially reveal a different relationship.
This is a 2021 ARVO Annual Meeting abstract.