# HCC Risk Adjustment Chart Review > Retrospective HCC chart review with page-cited proof for every supported diagnosis, and an explicit flag where the documentation does not support the code. Canonical page: https://medrecords.ai/solutions/hcc-risk-adjustment-chart-review/ --- [Home](https://medrecords.ai/) ›[Solutions](https://medrecords.ai/solutions/) ›HCC Risk Adjustment Review HCC RISK ADJUSTMENT CHART REVIEW ## Every diagnosis needs a page behind it. Most reviews don't show one. HCC risk adjustment chart review services abstract the supported diagnoses from a retrospective chart set and cite the page documenting each one. Codes with no supporting documentation are flagged rather than suggested, so the encounter submission your coders build rests on evidence that survives an audit. [Test a file ](https://medrecords.ai/test-a-file/?src=hcc-risk-adjustment-chart-review) [Book a demo](https://medrecords.ai/demo/) Click any row → the source page it cites Chart review · Case #IME-4812 documented Adams, Timothy · retrospective chart set 342 pp / 2 packets received logged 11 duplicate pages removed free p.140 — wrong patient quarantined Pages 342 Documents 27 Cited 100% ### The coding rules are published. So is the auditor who applies them. We do not build for a case type until we can name the standard that defines a correct output and the person who grades the file against it. In risk adjustment, both are already published. The standard and the referee The rulebook - The ICD-10-CM Official Guidelines for Coding and Reporting, including the requirement that a condition be documented as monitored, evaluated, assessed or treated in a face-to-face encounter - The CMS-HCC risk adjustment model for the applicable payment year, which maps diagnoses to HCCs - CMS risk adjustment data validation guidance, which sets what an acceptable record looks like - The plan's own compliance policy on retrospective review and on deletes as well as adds The referee - The CMS RADV medical record reviewer, who tests the same charts against the same guidance - The HHS Office of Inspector General, whose risk adjustment audits target unsupported diagnoses - The Department of Justice, where a one-directional review program becomes a False Claims Act exposure Coder hours per chart across a retrospective population, most of it spent reading Flat 10¢/page here, duplicates free Chart review · Case #IME-4812 cited 100% Packet342 pp / 2 packets · 11 deduplicatedlogged Supported diagnosesMEAT documented and page-citedcited Encounter validityProvider type, date, signaturecited Unsupported codesSubmitted with nothing behind itflagged Review delivered · cited 100% · no coding assignment How it works ### 3 steps between the charts and the submission. 01 #### Send the chart set The retrospective charts as you hold them — EHR exports, scanned encounter notes, specialist consults — in any format and any order. 02 #### We return the review, page-cited Every diagnosis the documentation supports, with the page showing it was monitored, evaluated, assessed or treated, and every unsupported code flagged in its own line. 03 #### Your coders decide Coding assignment and the submission decision, adds and deletes alike, stay with your certified coders and compliance team. The boundary, in writing ### Evidence for the coder, not a code on the claim. We abstract and cite what each chart documents, in both directions: diagnoses the record supports and codes the record does not. We do not assign codes, submit encounter data, or certify a diagnosis as compliant. Those are your certified coders' and compliance team's determinations. Flagging unsupported codes is not an optional module. A retrospective review that only ever adds is the pattern regulators look for, and a review that cites its evidence in both directions is the one that holds up. Audit trail · Case #IME-4812 exportable 08:12Packet received · 342 pp / 2 packetssystem 08:3111 pages deduplicatedsystem 08:44p.140 flagged — wrong patientsystem 08:54Chart review complete · review complete · both directions cited · cited 100%system 09:20Chart review reviewed · citations verifiedreviewer 09:26Chart review exported · encounter decisions made off-platformreviewer Every access logged · file deleted 30 days after delivery Why Medrecords AI ### The rules the platform never breaks. EVERY LINE CITED CASE #IME-4812 · ADAMS, T.342 pp 2/14 — ER visit, right knee p.4 4/18 — arthroscopic surgery p.61 p.140 — wrong patient quarantined Medrecords AI Read every page · cite every line #### HIPAA, under a signed BAA Every file is handled under our Business Associate Agreement, from the first byte. #### Never trains a model Your records are never used to train any AI model — ours or anyone else's. #### Every line cited If we can't cite it, we don't say it. Every sentence links to its source page. #### Deleted after delivery Files are deleted 30 days after delivery, with a full audit log of every access. Powered by the platform ### 4 capabilities behind every chart review. The chart review is 1 deliverable of the same platform that reads, sorts, and cites the whole record. [ #### Medical Records OCR Every chart read in full — EHR exports, scanned consults, handwritten encounter forms — nothing skipped, nothing guessed. IN ACTION · 342 pp / 2 packets → 27 documents, read in full ](https://medrecords.ai/product/ocr/) [ #### Medical Chronology Each patient's encounters build into a timeline, so a chronic condition's documentation across the year is visible at once. IN ACTION · first encounter → year end, 1 continuous timeline ](https://medrecords.ai/product/chronology/) [ #### Verifiable AI Citations If we can't cite it, we don't say it. Every supported diagnosis links to the page showing it was addressed. IN ACTION · every diagnosis → cited 100% ](https://medrecords.ai/product/citations/) [ #### Medical Summary Reports The review formats straight into the coding workflow — your template, your letterhead. IN ACTION · chart review → coder-ready worklist ](https://medrecords.ai/product/summaries/) FAQ ### HCC risk adjustment review, answered. No. We cite the documentation that supports or fails to support each diagnosis. Code assignment and the decision to submit an add or a delete stay with your certified coders and your compliance team. Yes, and it isn't optional. A retrospective review that only adds diagnoses is exactly the pattern OIG and DOJ scrutinize. Citing evidence in both directions is what makes the program defensible. It's the same evidence test, run before the auditor runs it. A diagnosis with a page-cited face-to-face encounter, valid provider type and legible signature is the one that survives RADV; one without is the one that gets extrapolated. No. Medrecords AI does not retrieve records from providers or facilities. You bring the records you already have — review starts in minutes from upload. Retrieval vendors take days; you can keep yours and still cut the review to minutes. ### Send 1 file. We'll tell you what we can read. No obligation. If the case type is buildable we'll scope it; if it isn't, we'll say so. [Test a file ](https://medrecords.ai/test-a-file/?src=hcc-risk-adjustment-chart-review) [Book a demo](https://medrecords.ai/demo/) [See every case type we process →](https://medrecords.ai/solutions/case-types/)