Myopia is a refractive error in which the eye is too long for the focussing power of the lens and so the image is brought to focus a little way in front of the retina which results in a blurred image for the affected person. In most cases this can be corrected by spectacle or contact lenses and more recently by laser surgery and many billions are spent annually across the world on these interventions. However these treatments do not reduce the excess risk for the myopic individual of other serious eye conditions such as glaucoma, cataract, retinal detachment and macular degeneration and myopia remains a significant cause of blindness in the UK, for example myopia was the 4th most frequent cause of registration for blindness in Scotland in 1990's (1).

In the last century there has been a rapid increase in the prevalence of myopia, particularly in far eastern countries where now the condition is said to have reached epidemic proportions as the majority of young people are myopic in recent surveys, for example over 80% of children leaving school in Taiwan and in Singapore, in the 1990's are myopic (reviewed in 2). In the UK, Australia and US recent data suggest lower prevalences but there is evidence of a cohort effect so that prevalences are increasing with successive generations. Although much data suggests a strongly heritable component to myopia (3), the recent marked increases in prevalence are likely to be driven by environmental exposures and propensity for or intensity of reading (4,5), decreased time outside (6), increased urbanization (7) and changing dietary patterns (8) have all been implicated in observational studies. The problems of confounding and bias however, cause major difficulties for researchers when trying to hypothesize or test causal mechanisms leading to myopia.

By contrast with the epidemiological literature, in experimental animal studies there is strong evidence that manipulation of visual input will produce excessive eye growth leading to myopia, for example by blocking all visual input, or just patterned input, or by using defocussing lenses (9). The myopia-causing effects of imposed defocus can be inhibited by short bursts of light (10) . Dopamine is present in the mammalian retina and is released in response to light exposure (11). Drugs which mimic the action of dopamine prevent the development of myopia in form-deprived eyes (12). In a recent paper it was hypothesised that dopamininergic mechanisms may mediate the protective effect of time spent outside in humans (10) and this question was a "hot topic" at the July 2008 International Myopia Conference (13)

Within ALSPAC, we have already published data that shows the expected associations between increased risk of myopia and increased exposure to reading (4). Unpublished data also show a protective effect of increased tine outside being (3+ hrs vs less, on school summer days) associated with a lesser risk of myopia at 11yr, after adjustment for sex, ethnicity, parental myopia and time spent reading (adjusted OR 0.64; 95% CI 0.49 to 0.86, p = 0.001). Also within ALSPAC work has already been published describing SNPs relating to the dopamine system (in the genes for the enzymes COMT and MAO-A) and examining associations between functional variants in these genes and characteristics of the participants including the children's IQ at 8 (14,15), their emotional status at 6 - 7 years (16) and their hyperactivity at 4 and/or 7 years (17).

We hypothesize that dopaminergic mechisms may be involved in myopia development in humans. We would therefore like to test the following related hypotheses, using data already available in ALSPAC:

1 - that functional variants in the COMT gene, and a polymorphism in the MAO-A gene, are assocated with risk for myopia at 11 and/or 15. The SNPs are rs2075507, rs6269, rs4818, rs4680, rs165599 and rs737865 . We will test these individually and also use the "Nackley haplotype" (18)which gives 6 diplotypes with progessively lesser levels of activity of the COMT gene, therefore greater dopamine availability. We will also test the variant in the MAOA (promoter VNTR AJ004835).

2 - that one or more alleles in the above SNPs mediate (or tag a functional variant that mediates) the observed protective effect of spending more time outside, on myopia development. This will be tested even if there is no main effect in (1) above.

All analyses will include stratifying by sex to look for interaction as dopamine expression/activity varies with sex in several reports.

The results of these analyses will be useful as they will be relatively unaffected by confounding and will provide the first data in humans regarding the potential importance of dopaminergic mechanisms for myopia development in a population of children.