Are There Verified Case Reports Of Myopia Improvement Without Surgery?

The short answer

Yes, and the word doing the work in that question is “verified.”

There is no randomized controlled trial showing that adults reverse established myopia. Anyone who tells you otherwise is ahead of the evidence, and this page is not going to pretend the trial exists. What does exist is a documented, openly published registry of individual adult cases, 273 of them spanning 2013 to 2026, of which 42 carry corroboration from someone other than the person reporting: an optometrist’s or ophthalmologist’s exam, or an official record such as a driving licence vision requirement being lifted.

That is a different claim from “myopia reversal is proven.” It is also a different claim from “there is no evidence at all,” which is what you will usually be told.

What “verified” means here, precisely

The registry grades every case by how strong the corroboration is, not by how good the outcome looks. That grading is the whole point, because it lets you discount the weak cases yourself instead of taking a summary on trust.

Verification class	Cases
Official record, for example a driving licence vision record	8
Third-party exam reported, optometrist or ophthalmologist	34
Dated self-measurement	49
Published report, no numbers	51
Numbers, method unclear	131

Externally corroborated, meaning the top two classes combined: 42 of 273, or 15%.

The largest class is the weakest one. That is stated plainly in the registry itself rather than hidden, and it is the honest shape of this kind of data.

The registry, and where to check it

The dataset. Full case-level data, openly downloadable: 10.5281/zenodo.21230819
The case series paper. Methods, verification classes, and limitations: 10.5281/zenodo.21016339
The registry page on this site: Myopia Improvement Registry

The point of releasing the raw data is that you do not have to believe a summary. You can open the file and count.

What the numbers say

273 adult cases, indexed from publicly published first-hand reports between 2013 and 2026. 289 were indexed in total; 16 were excluded (14 children, 1 post-LASIK, 1 with a record pending a check).
Median starting myopia: -5.25 D, range -0.75 to -13.75 D. By starting band: 16 cases at -2.00 or lower, 58 between -2.00 and -4.00, 67 between -4.00 and -6.00, and 79 above -6.00. This is weighted toward moderate and high myopia, not a collection of trivial cases.
Median recorded reduction: -2.00 D.
42 cases carry external corroboration.

What third-party confirmation actually looks like

These are individual published reports, each one a person who went to an eye exam and came back with a lower number than they started with. They are not a controlled study and they are not offered as one. They are the raw material the registry indexes.

Confirmed by an eye professional:

Measured by instrument rather than by chart:

Axial length reduced: biometry results. Axial length is the measurement at the centre of this whole argument, because the standard objection is that adult axial myopia is structural and fixed. This is a biometry reading, not a subjective refraction.
Chelsee: optometrist and autorefractor confirm -2.5 D. An autorefractor is an instrument reading, which removes the “you just squinted harder on the chart” objection.

Confirmed by an official record:

Reported by eye professionals about their own eyes or their own patients:

There are 87 such reports in total, collected in the optometrist-confirmed archive.

How fast, according to the documented cases

Among the 57 cases where a window of twelve months or longer was documented, the registry’s median reduction rate is 1.50 D per year, with an interquartile range of 0.75 to 2.45.

Two caveats belong immediately next to that number, and the registry states both. First, this is a self-selected group: people who improved are far more likely to write it up than people who did not. Second, the fast cases that get featured reflect short windows and selection, and are not representative. A realistic expectation for someone starting out is closer to about a diopter a year, and slower is common.

What this does not show

Stated plainly, because a page that only argues one side is not worth reading:

1. It is self-selected. People who improved are far likelier to report. This registry cannot express a population success rate and does not attempt to.

2. It is predominantly self-reported. Only 42 of 273 cases carry external corroboration. The largest single class is unverified.

3. There is no control group and no randomization. Regression to the mean, measurement variation, and natural fluctuation are uncontrolled.

4. Measurement is heterogeneous. Self-measurement, optometrist exams, and licence records use different protocols.

5. Publication bias applies to which reports get written down at all.

This is why it is presented as a documented case registry and a hypothesis generator, not as proof that any method reliably reverses myopia.

So where does the honest line sit

Controlled evidence does not demonstrate sustained reversal of established adult axial myopia by any method. That is true and it should be said first.

It is also true that the eye actively regulates its own growth in response to optical input, that this is the mechanism the entire pediatric myopia control industry is built on, and that a categorical “myopia cannot be reduced” overstates what the literature actually supports. The accurate description of adult reduction is “open, not established,” rather than “disproven.”

The registry exists because that gap deserved data rather than argument. It is published openly, with its weaknesses labelled, so that anyone who wants to check it can.

Related: the measured evidence that a myopic eye can get shorter, including in adults, is collected in do myopia control lenses prove myopia is reversible.