# Missing Medical Records Identification > Missing medical records identification cross-references treatment history against what was produced, flagging visits, providers, and dates that should exist. Canonical page: https://medrecords.ai/product/missing-records-identification/ --- 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/) › — Missing Records Identification MISSING MEDICAL RECORDS IDENTIFICATION ## The gap in the file, found before opposing counsel finds it. Missing medical records identification cross-references the treatment history against what was actually produced and flags the visits, providers, and date ranges that should exist but don't. Each flag is cited to the evidence that implies it, so the gap in the file is found — and documented — before opposing counsel finds it. [Test a file ](https://medrecords.ai/test-a-file/?src=product-missing-records-identification) [Book a demo](https://medrecords.ai/demo/) Adams, Timothy — right knee · Case #IME-4812 342 pages · 2 packets Expected record — Evidence — Status Specialist consult Referral cited p.418 MISSING PT visits 8+ 7 of plan produced p.412 GAP MRI report Order + billing line p.129 PRODUCED Prior-injury records History mentions 2020 care MISSING Each flag names **the page that implies the missing record** — judge it yourself. Why it matters An incomplete file reads fine until someone else completes it. The gap analysis runs **before your opinion is on the line** , not after. ### The history says what should be there. Every produced page contributes to a reconstructed treatment history: who treated, when, and what was ordered next. That history is then cross-referenced against the actual production. A referral with no consult note, a billing line with no visit, a follow-up ordered but never appearing — each becomes a candidate gap. Referrals, orders, billing lines, and refills all checked for follow-through Flags, not verdicts — you confirm what's truly missing Signals that imply a record REFERRAL "Referred to orthopedic specialist" — but no specialist note in the file BILLING A charge for a visit whose clinical note was never produced FOLLOW-UP "Return in 4 weeks" — and the timeline goes quiet instead HISTORY Intake mentions prior care at a facility absent from production Every signal carries its **page citation** . Gap table · export request-ready Provider — Expected range — Evidence Orthopedic specialist after referral Referral note p.418 Physical therapy after 4/02 Plan of care p.412 County Medical prior care Intake history mention Drops straight into a **gap letter or follow-up request** — no retyping. ### From flag to request-ready gap table. Confirmed gaps export as a table the requesting team can work straight down: provider, expected date range, the evidence behind the gap, and its citation. It feeds gap letters and follow-up requests directly — and when Evidence Request Automation ships, it will feed that queue too. Gap letters drafted from the table, evidence cited New productions re-checked automatically against the same history ### "The file is complete" — as a documented finding. Completeness is worth as much as any gap. When the cross-reference comes back clean, that's a citable statement for the report: the treatment history was checked against production and no unexplained gaps remained. Far stronger than "nothing jumped out." Completeness statement with the method documented Resolved and dismissed flags kept, with reasons, for the audit trail Completeness check documented Treatment history reconstructed 342 pages read Cross-referenced against production 2 packets Flags reviewed and resolved reasons logged No unexplained gaps remain FINDING Citable in the report, defensible in the deposition. Timeline · treatment window gap detected Quiet window after ordered follow-up "Return in 4 weeks" documented — then nothing until the next packet begins. Flagged with the ordering note cited. A flag is a question for a human, never a silent conclusion. The standard ### Flagged and cited — never guessed. A missing-record claim you can't support is a liability of its own. Every gap flag traces to the page that implies it, so the assertion "records were not produced" arrives audit-grade and legally defensible — with the evidence attached, and the final judgment yours. [See Verifiable AI Citations ](https://medrecords.ai/product/citations/) ### From produced file to gap analysis. 3 steps, run on every file — and re-run when new records arrive. 01 Upload what you have The produced file is read in full — notes, billing, referrals, and orders included. 02 History vs. production The reconstructed treatment history is cross-referenced against what was actually produced; mismatches become cited flags. 03 Export the gap table Confirm or dismiss each flag, then export the request-ready table or the completeness finding. ### Who hunts the gaps. The 2 teams for whom an incomplete file is a professional risk. [ Law firms Find the hole in your production before opposing counsel builds a cross around it. For law firms ](https://medrecords.ai/solutions/law-firms/) [ Medical evaluators Know what you weren't given before the exam — and say so in the report, with evidence. For evaluators ](https://medrecords.ai/solutions/medical-evaluators/) FAQ ### Missing records identification, answered. The platform builds the treatment history from what's in the file — referrals, provider mentions, billing lines, medication refills, scheduled follow-ups — and cross-references it against the documents actually produced. Anything the history implies but the file doesn't contain is flagged as a candidate gap, with the evidence cited. A referral to a specialist with no note from that specialist. A billing line with no matching clinical note. A follow-up ordered but never appearing. A medication refill implying visits in an undocumented window. Each flag names the page that implies the missing record, so you can judge it yourself. An exportable table of every suspected gap: provider, expected date range, the evidence it rests on, and its page citation. It's formatted so the person sending record requests can work straight down the list — and it feeds gap letters and follow-up requests without retyping. It can document completeness as a finding: the treatment history was cross-referenced against production and no unexplained gaps remained. That's a defensible, citable statement for a report or an audit — stronger than "we didn't notice anything missing." No — it identifies what's missing and prepares the gap table; sending requests stays with your team. Request tracking with tiered follow-up escalation is a separate capability, Evidence Request Automation. ### Related capabilities [ Supplemental Record Review When the requested records arrive, they land pre-deduplicated and flagged where they matter. Explore ](https://medrecords.ai/product/supplemental-record-review/) [ Evidence Request Automation Tracking and tiered escalation for the requests your gap table generates. Explore ](https://medrecords.ai/product/evidence-request-automation/) [ Medical Chronology The timeline where treatment gaps first become visible, event by cited event. Explore ](https://medrecords.ai/product/chronology/) ### Find out what your file is missing. Upload a single file and get the gap analysis back — every flag cited to the page that implies it. Handled under our BAA; never used to train a model. [Test a file ](https://medrecords.ai/test-a-file/?src=product-missing-records-identification) [Book a demo](https://medrecords.ai/demo/) [All capabilities →](https://medrecords.ai/product/)