Age of myopia onset in a British population-based twin cohort
School-age myopia is becoming more common in Asia and North America; data from the United Kingdom has suggested a significant amount of myopia develops after the age of 17 years. Age of spectacle wear has been used as a proxy of myopia severity in a recent large genome-wide association study. The purpose of this study was to examine the age of onset of spectacle wear in a large British twin cohort, to examine the reliability and reproducibility of self-reported age of onset as a proxy measure of myopia severity, and to see if there is evidence in the UK of a rising prevalence of myopia.
Non-cycloplegic autorefraction was performed on over 6000 subjects from the TwinsUK cohort, a large, well-characterized volunteer cohort of British, predominantly Caucasian female twins, between 1998 and 2010. Questionnaires asking age of first spectacle wear were conducted in 2003 and 2008. Myopia was defined as worse than or equal to −1.00 Dioptres, and adult onset myopia as occurring on or after the age of 17 years.
Autorefractive data was available on 6097 participants at a mean age of 53 years. The mean S.E. was −0.36 D (S.D. 2.67, range −25.13 to +9.38). 1705 subjects (28%) were myopic with a mean refractive error of −3.54 (S.D. 2.51, range −25.13 to −1.00) and the median age of first glasses wear was 15 years (mean 18.4 years, S.D. 12.24, range 0–74). Of those who provided an age at which they first wore glasses in both questionnaire sources (n = 628), there was median difference in response of 0 years (S.D. 7.18, mean 0.7, maximum 53). A statistically significant cohort effect for increased myopia prevalence across a range of age groups between 1998–1999 and 2008–2010 was identified, with myopia prevalence increasing from 27% to 34% in those aged 50–54 and from 16% to 32% in those aged 55–59.
Almost half the myopes in this UK-based population wore glasses after the age of 17; further research into adult-onset myopia is required. Although self-reported age of glasses is reproducible and reflects severity, it only explains approximately 15% of the variance of spherical equivalent, so is a rough proxy of refractive error, but still may be useful in large-scale population studies without access to refraction. We have demonstrated a significant cohort effect for increased myopia prevalence in the UK population over a 10-year period.