# Co-Mingled Medical Records Detection > Co-mingled medical records detection that flags pages belonging to another patient and quarantines them — preventing a PHI incident instead of reporting one. Canonical page: https://medrecords.ai/product/co-mingled-records-detection/ --- NewWe now read the actual MRI and CT imaging — not just the radiologist's 1-page report. [See how →](https://medrecords.ai/product/imaging/) [Home](https://medrecords.ai/) ›[Product](https://medrecords.ai/product/) › — Co-Mingled Records Detection CO-MINGLED MEDICAL RECORDS ## The wrong patient's pages, caught before anyone reads them. Co-mingled medical records detection from Medrecords AI catches pages belonging to another patient before anyone reads them. A misfiled page is a privacy incident waiting to happen — a reportable event — so suspect pages are flagged and quarantined before analysis begins, and every incident prevented is still documented. [Test a file ](https://medrecords.ai/test-a-file/?src=product-co-mingled-records-detection) [Book a demo](https://medrecords.ai/demo/) Adams, T. — right knee · Case #IME-4812 342 pages p.140 — demographics don't match the case patient p.139 — Progress note Adams, T. verified p.140 — Lab report, different patient QUARANTINED p.141 — Progress note Adams, T. verified p.142 — Imaging order Adams, T. verified 1 page quarantined · excluded from every downstream output p.140 The wrong-patient page, quarantined at intake Before analysis Ahead of the chronology, chat, and every report Every event logged Detection, quarantine, and release all leave a trail ### How a stranger's page gets caught. Every page is checked against the case patient: name, date of birth, identifiers, and clinical continuity. A page that names someone else — or describes a body part, medication, or history that contradicts the rest of the file — is flagged with the mismatch stated. Demographic mismatch detection on every page Clinical-consistency signals catch pages with no name at all Mismatch report · p.140 Patient name on page — does not match — p.140 Date of birth — does not match — p.140 Clinical continuity — left shoulder in a right-knee file The flag states **why** the page doesn't belong — not just that it doesn't. Quarantine state p.140 — quarantined — awaiting decision Hidden from chronology, search, chat, reports, and exports. Content masked pending review. Chronology — page excluded Search & chat — page excluded Reports & productions — page excluded ### Quarantine, before anyone reads it. A flagged page is pulled aside immediately: it never enters the chronology, never surfaces in search or chat, and never lands in a report or production. Reviewers see that a page was quarantined and why, without the other patient's PHI spreading any further. Excluded from every downstream output automatically Release requires an explicit human decision ### An incident prevented is still documented. Each quarantine writes an audit entry: the page, the stated reason, the timestamp, and the reviewer's eventual decision. If compliance asks what happened, the answer is a log, not a reconstruction from memory. Exportable quarantine log per case Release or removal — a human decision, recorded either way Quarantine log · p.140 Detected at intake — logged Quarantined — reason stated — logged Reviewer decision — removed from case — logged The full log exports with the case — **audit-grade** from detection to decision. Quarantine · p.140 Different patient detected demographics mismatch Excluded from analysis automatic Reviewer decision pending human-gated The other patient's PHI never reaches a reviewer's screen. PHI incident prevention ### One misfiled page is a reportable event. Co-mingled pages turn up in real productions more often than anyone admits: a fax loaded into the wrong chart, a scanner batch that ran long. Caught downstream, that's a privacy incident with notification duties. Caught at intake, it's a log entry. Audit-grade review starts with knowing every page belongs. [See Audit Trail & Chain of Custody ](https://medrecords.ai/product/audit-trail-chain-of-custody/) ### From upload to a file you can trust. Identity checks run at intake, before a single page is analyzed or shown. 01 Upload the production Files from providers, payers, and prior counsel go in as received — misfiled pages and all. 02 Every page identity-checked Name, date of birth, identifiers, and clinical continuity are verified against the case patient. 03 Quarantine and decide Mismatched pages are held out of every output until a human releases or removes them — each step logged. ### Who catches misfiled pages with it. Anyone who merges records from multiple sources into one file — which is everyone. [ IME & QME evaluators An evaluator's report built only on the right patient's history. For medical evaluators ](https://medrecords.ai/solutions/medical-evaluators/) [ Law firms No stranger's PHI in your production, exhibits, or demand package. For law firms ](https://medrecords.ai/solutions/law-firms/) [ TPAs & case management Files from many sources, checked before they merge into one claim. For TPAs ](https://medrecords.ai/solutions/tpas/) [ Insurance carriers Privacy incidents prevented at intake and documented for compliance. For carriers ](https://medrecords.ai/solutions/insurance-carriers/) FAQ ### Co-mingled records, answered. Every page is checked against the case patient on 2 axes: identity signals — name, date of birth, record identifiers — and clinical continuity, meaning whether the page's content is consistent with the rest of the file. A left-shoulder operative note inside a right-knee case gets flagged even if no name appears on the page. It is pulled out of the analysis pipeline immediately: excluded from the chronology, search, chat, reports, and any export or production. Reviewers see that a page was quarantined and the stated reason, but the content is held back so the other patient's PHI stops spreading. A co-mingled page that reaches a report, a production, or opposing counsel is a disclosure, with the notification duties that follow. Detection at intake means the page never reaches a downstream output, so what would have been an incident becomes a documented catch. Only by an explicit human decision. If a reviewer determines the page does belong to the case patient — a maiden name, a transcription error — they can release it, and the release is logged with the reviewer and reason. Nothing is released automatically. Every step writes an entry: detection, the stated mismatch, the quarantine, and the eventual release or removal, each with a timestamp and actor. The log exports with the case, so compliance gets evidence rather than recollection. ### Related capabilities Adjacent features on the same platform — every output source-linked and cited to page. [ Medical Record Deduplication Duplicate pages across merged files collapsed to one canonical record. Explore ](https://medrecords.ai/product/medical-record-deduplication/) [ PHI Redaction & Production Prep PHI located and redacted before a document leaves the platform. Explore ](https://medrecords.ai/product/medical-records-redaction/) [ Audit Trail & Chain of Custody An immutable log of every action taken on the file. Explore ](https://medrecords.ai/product/audit-trail-chain-of-custody/) ### Check whether every page belongs. Upload one production and see the identity check run page by page. Handled under our BAA; never used to train a model. [Test a file ](https://medrecords.ai/test-a-file/?src=product-co-mingled-records-detection) [Book a demo](https://medrecords.ai/demo/) [All capabilities ](https://medrecords.ai/product/)