Half the World Is Projected to Be Short-Sighted by 2050. The Genes Did Not Change

Myopia is strongly heritable and rising faster than any gene pool can move. A polygenic score built from 30,217 variants shows where your DNA sits — and the paper it comes from is mainly about why such scores travel badly between populations.

In 2000, an estimated 1,406 million people were short-sighted — about 22.9% of the world. The projection for 2050 is 4,758 million people, or 49.8%, with high myopia rising from 2.7% to 9.8% of the global population (Holden et al., Ophthalmology, 2016, PMID: 26875007).

Fifty years is nothing in evolutionary time. Whatever is driving that, it is not a change in human genetics.

Two true statements that sound contradictory

Myopia is strongly heritable. Children of short-sighted parents are considerably more likely to become short-sighted. Twin studies and large GWAS both find substantial genetic contribution.

Myopia prevalence has roughly doubled in a generation in some populations. In parts of urban East Asia, rates among young adults have reached levels far above those of their grandparents.

Both are correct, and reconciling them is the whole point. Heritability describes how much of the variation within a population at a given time tracks genetic variation. It says nothing about what happens when the environment shifts for everybody at once.

The environment did shift. Near work rose, and — the factor with the strongest evidence behind it — time spent outdoors fell. Outdoor light exposure in childhood appears protective, and it is the lever that intervention studies have most consistently found to matter.

Your genotype influences where you sit relative to your peers. It does not set the level for the whole cohort.

What the score actually is

The app uses PGS001994, a published polygenic score from the PGS Catalog, built from 30,217 variants.

Its source is worth knowing, because it is unusually candid. The score comes from a study whose subject is "Portability of 245 polygenic scores when derived from the UK Biobank and applied to 9 ancestry groups from the same cohort" (Privé et al., American Journal of Human Genetics, 2022, PMID: 34995502).

That paper exists to quantify a known problem: polygenic scores built in one population lose accuracy when applied to another, and the loss grows with genetic distance from the training set. A score developed in a largely European-ancestry biobank performs measurably worse in other groups.

This is why the app reports your percentile within an ancestry-matched reference population rather than against one universal distribution. It is the correct handling of the limitation, and it is unusual to see it done explicitly.

What it does not do

GenePlaza's own description is blunt about this, and it deserves repeating rather than softening: the app is "for education and entertainment only, it is not a medical or diagnostic test."

It will not tell you your prescription, whether your child will need glasses, or whether your eyesight will deteriorate. Myopia is diagnosed by an optometrist with a refraction test, which takes ten minutes and is far more informative than any genome.

If you or your child have vision concerns, that appointment is the answer — not this.

What the app reports

The Myopia App gives you a percentile, an interactive distribution chart and a printable report showing where your DNA places you for short-sightedness.

As with every app on the platform, that percentile is a position: where you would have fallen within the reference population had you been part of that analysis. Interesting, honestly bounded, and not a forecast about your eyes.