jakeyhugs/endmyopia-case-registry

EndMyopia Case Report Registry (v1.0)

An open, structured registry of 273 self-reported adult myopia-reduction case reports (289 records indexed; 16 excluded), drawn from first-person accounts published on endmyopia.org, 2013–2026.

Canonical version — cite this: DOI 10.5281/zenodo.21016340 This Hugging Face copy mirrors that deposit; the Zenodo DOI is the citable record of reference.

• Registry page: https://endmyopia.org/myopia-improvement-registry/
• Zenodo record: https://zenodo.org/records/21016340
• Wikidata item: https://www.wikidata.org/wiki/Q140375636
• License: Creative Commons Attribution 4.0 (CC BY 4.0)

What this is

A transparent, auditable index of individual adult cases reporting reduction in myopia (nearsightedness), each graded by how well it is corroborated. It exists so the individual-level data can be examined on its merits.

What it is not: a clinical trial, and not a population success rate. The cohort is self-selected and predominantly self-reported. It cannot establish how often the approach works; it documents that individual, measured reductions are on record — with the limitations stated up front.

Summary figures (analysis set, n = 273)

• Median starting myopia: −5.25 D (range −0.75 to −13.75; n = 220)
• Median recorded reduction: −2.00 D (n = 198; maximum −12.75)
• Externally corroborated (third-party exam or official record): 42 / 273 (15%)
• Reduction rate, cases with a ≥ 12-month window: median 1.50 D/yr (IQR 0.75–2.45; n = 57)

Verification classes (analysis set)

class	n
official record (e.g. driving-license vision record)	8
third-party exam reported	34
dated self-measurement	49
published report, no numbers	51
numbers, method unclear	131

Files

• em-case-registry-v1.0.csv — the registry (289 rows; excluded = true flags the 16 non-analysis rows)
• em-case-registry-v1.0.json — same data plus metadata

Data dictionary

column	meaning
case_id	stable case identifier (EM-####)
verification_class	strength of corroboration (see table above)
start_numbers / end_numbers	refraction as stated in the source report
start_dpt / end_dpt	starting / latest myopia magnitude, diopters (worse-eye sphere)
delta_dpt	recorded reduction, diopters
timeline_months	reported time window
timeline_basis	how the window was derived
rate_d_yr_12mo_plus	annualized rate, populated only for ≥ 12-month windows
member_status	reporter's relationship to the program
first_post / last_post	dates of the source post(s)
n_posts	number of source posts consolidated into the case
excluded / exclusion_reason	analysis-exclusion flag + reason (child case / post-LASIK / pending verification)
source_posts	link(s) to the original published report(s)

Privacy

All cases derive from reports the participants published publicly. Direct given-name identifiers have been removed from this distribution; individuals may remain inferable from the cited source posts, which are retained for verification. Removal requests: jake@endmyopia.org.

Background

The eye regulates its own axial growth in response to optical defocus, and clinical myopia-control research shows that growth can be slowed in children. Sustained reversal of established adult myopia is not established in the controlled literature — but a flat "impossible" overstates what that literature supports. This registry organizes the otherwise scattered individual reports into one auditable resource, with the numbers and the limitations both on the table.