James Loughman is a Professor of Optometry and Vision Science at Technological University Dublin and Director of the Centre for Eye Research Ireland, a dedicated ophthalmic research and clinical trials centre focussed on major causes of vision loss, in particular, myopia. James is also co-founder and CEO of Ocumetra, a data analytics company established to power the evolution of refractive error and ocular health management using big data analytics, communication infographics, customised device integration value-add solutions and insightful business analytics.
This abstract was presented at the 2022 ARVO Annual Meeting, held in New Orleans, USA, April 23 – 27, 2022. Download the original presentation slides (pdf).
Asian and Western refractive centile curves from meta-analysis of population refraction data
James Loughman, Kathryn J. Saunders, K Siofra Harrington, Daniel I. Flitcroft.
Purpose:
To generate childhood refractive centile curves for Asian and Western populations in order to facilitate evidence-based myopia control.
Methods:
35,645 refraction measurements from 8 large population-based studies were included (12,965 European/North American eyes, and 22,680 Asian eyes). An individual meta-analysis of right eye SER was performed by region, age and gender. Quantiles were generated by age group and gender. Combined curves for each calculated centile were generated by weighted cubic spline interpolation according to the number of participants in each age group/study.
Results:
Western and Asian population centiles differed at all ages (Figure 1). At age 6, median refraction was 1.33D and 1.34D in Western boys and girls, respectively and 0.25D and 0.24D in Asian boys and girls. Western children at the 97th centile were hyperopic at age 6 (0.11D boys; 0.18D girls). In Asia, 6-year-olds at the 97th centile were significantly more myopic (-1.71D boys; -1.77D girls). In the West, the median refraction did not reach the threshold for myopia (≤-0.50D) by age 18. In Asia, median refraction met the myopia threshold by age 10 in boys and age 9.5 in girls. Gender differences were most marked in hyperopes in the West, with boys 0.60D more hyperopic at age 7.5 at the 3rd centile. At age 10.5, girls were 0.22D more myopic at the 97th centile, but by 18 the gender difference was <0.05D at all centiles. Median refraction varied less than 0.02D between sexes at all ages in the West. In Asia, gender differences were minimal (<0.20D) below age 11 but increased with age and were greatest in myopes. The gender difference in median refraction increased from age 6 to 15, with girls 0.55D more myopic by age 15, reducing to 0.40D by age 18.
Conclusions:
Population-based centile analysis offers the prospect of monitoring the treatment efficacy of myopia control interventions outside a clinical trial but must be gender and region-specific. Historic data derived from the period before myopia treatments were introduced represent a valuable natural history resource on myopia progression. Pooling data from the large number of existing surveys provides a mechanism to standardise reference centile charts relevant to myopia management.
