Editorial & evidence standards.

Retina.blog is designed around inspectable evidence, clear uncertainty and transparent authorship.

Source hierarchy

We prioritize systematic reviews and meta-analyses where appropriate, randomized or controlled trials, prospective and observational research, peer-reviewed primary studies, clinical-trial registries, and authoritative institutions. Preprints may be covered when newsworthy but are clearly labeled as not yet peer reviewed.

Evidence labels

Established indicates mature evidence or accepted use. Supported indicates meaningful evidence with important limitations. Emerging indicates promising but incompletely validated work. Preliminary is used for early-stage findings. These labels are editorial summaries, not clinical recommendations.

Research data model

Production content should retain DOI/PMID identifiers when available, study type, peer-review status, publication date, limitations, last evidence-review date, authorship and reviewer information.

Corrections & updates

Article dates should reflect genuine publication or substantive revision. Evidence reviews should be updated when materially important new research changes the interpretation.

AI assistance

AI may assist research organization, drafting and production. It is not treated as a medical authority. Claims should remain traceable to credible sources and receive human editorial review before publication.