Editorial & data integrity policy

Last updated: 5 September 2026

SameCase publishes three distinct kinds of content: clearly labeled illustrative teaching cases from SameCase Editorial, de-identified clinical cases contributed by verified doctors, and account-free patient case descriptions. This page explains how each kind is sourced, labeled, protected, and corrected.

Illustrative teaching cases

A page labeled "Teaching example" is a constructed educational scenario, not a real patient record and not a clinician testimonial. SameCase Editorial writes these examples to demonstrate the structure and educational purpose of the library while the verified-doctor corpus grows. Public pages include the constructed narrative, workup, resolution, and teaching point. Diagnosis-forward titles make the subject discoverable; the scenario remains prominently labeled illustrative throughout. Structured demographic and symptom-list facets remain inside the member experience.

Teaching examples are always attributed to SameCase Editorial, carry the same label on library cards and detail pages, and include machine-readable educational labeling. We do not create endorsements, comments, or named clinicians for them, and illustrative outcomes are never presented as observed patient outcomes. These pages are excluded from any count described as clinician-contributed or real-world cases. Unless a page names a real reviewer and links a supporting profile, no named clinician reviewer has been assigned; SameCase never invents reviewer attribution.

Who can publish

Apart from the clearly labelled SameCase Editorial teaching examples described above, only clinicians whose medical registration number (NMC / state / national register) has been reviewed can publish cases, comment, or respond to patient questions. Public doctor profiles are not proof of identity: each profile is privately attributable to a currently verified account. Pseudonymous display is allowed, but every pseudonym is backed by completed identity and licence checks. Patients can submit their own case description without any account — those submissions are matched against the corpus but are never presented as clinical guidance.

Case standards

Clinician-contributed cases must be real, de-identified, and outcome-linked: presentation, relevant workup, and the eventual diagnosis or resolution. Cases must not contain any patient-identifying information — our submission flow runs an identifier scanner, and the terms of use make de-identification a condition of publishing. Cases that are promotional, fabricated, or identifying are removed and the account reviewed.

De-identification pipeline

Patient text passes an identifier scanner in the browser (names, phone numbers, emails, record numbers, exact dates, facility names) so identifying details can be removed before anything is sent — and the server runs a second scrub before storage. These automated checks cannot confirm de-identification, and residual re-identification risk remains. A patient report is never public. An opt-in only queues the individual narrative for human privacy review; it remains quarantined unless approved, after which only currently verified doctors can see it. Thresholded aggregate patterns remain separate from individual access.

Use of AI

Matching is text-similarity search across the de-identified corpus — it is a retrieval system, not a diagnostic model. The "questions for your doctor" shown to patients are generated from patterns in matched solved cases and are always framed as discussion prompts, never as diagnoses or treatment advice. No content on SameCase is auto-published by AI. Clinician-contributed cases are written and owned by verified doctors; illustrative cases are separately authored and labeled by SameCase Editorial.

Protecting the corpus

We allow search engines and AI answer engines to index and cite our public pages (that is how people find SameCase), but we block AI training crawlers from harvesting the case corpus through robots.txt and application enforcement. Training crawlers may read the marketing and editorial surfaces; requests to protected case or profile paths receive a hard 403 response. Our machine-readable policy is search=yes, ai-input=yes, ai-train=no.

No ads, no data selling

SameCase runs no advertising, uses only aggregate cookieless Cloudflare Web Analytics, and does not sell data — see the privacy policy. The platform is free; planned future revenue is optional premium analytics for institutions, built on aggregates, not on personal data.

Corrections and takedowns

If you believe a case contains identifying information, is inaccurate, or should not be on the platform, contact us with the case ID. Identifying-information reports are treated as urgent: the case is hidden while reviewed. Factual corrections are applied to the case with the contributing doctor, and material changes are visible in the case history.

Medical disclaimer

All content on SameCase is educational. It is not medical advice, not a diagnosis, and not a substitute for consultation with a qualified clinician who can examine you. A similar presentation can have a different cause. If symptoms are severe or worsening quickly, seek urgent medical care.