# Utilization Review & IMR Record Packets > Record packet assembly for utilization review and independent medical review. Indexed and cited against the applicable treatment guideline and formulary. Canonical page: https://medrecords.ai/solutions/utilization-review-record-packet/ --- [Home](https://medrecords.ai/) ›[Solutions](https://medrecords.ai/solutions/) ›[Workers' Compensation](https://medrecords.ai/solutions/workers-comp/) ›Utilization Review Record Packets UTILIZATION REVIEW RECORD PACKETS ## The IMR packet, built to the guideline section it turns on. Utilization review record packets assemble one disputed treatment request and its fifty to three hundred pages of support: indexed, deduplicated, and cited to the MTUS section and formulary entry the dispute turns on, before the file reaches the physician reviewer. We organize and cite the record. The reviewer decides medical necessity. [Test a file ](https://medrecords.ai/test-a-file/?src=utilization-review-record-packet) [Book a demo](https://medrecords.ai/demo/?src=utilization-review-record-packet) Click any row → the source page it cites Record packet · Case #IME-4812 indexed Adams, Timothy · IMR record submission 342 pp / 2 packets received logged 11 duplicate pages removed free p.140 — wrong patient quarantined Pages 342 Documents 27 Cited 100% ### The guideline is published. So is the scoreboard. 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. California utilization review has both, in statute and in regulation, down to the guideline edition. It also has something almost no other case type has: an annual, public result. The state publishes how often the reviewer disagreed with the denial, so the quality of what was submitted is measured in the open. The rulebook and the referee The rulebook - Labor Code §4610, the utilization review statute, and §4610.5–4610.6, which govern independent medical review of a UR denial or modification - 8 CCR §9792.6–9792.12, the utilization review regulations - The MTUS at 8 CCR §9792.20 et seq, incorporating the ACOEM Practice Guidelines and the MTUS Drug Formulary - ODG where the MTUS is silent The referee - Maximus Federal Services, the sole California independent medical review organization, contracted through September 2027 - Maximus issues a written determination carrying an explicit guideline citation - DWC publishes the overturn rate annually: 10.2% in 2025, down from 12.7% in 2024 - The best public scoreboard in workers' compensation, and it grades what reached the reviewer Roughly 152,000 California IMR determinations a year at about $400 each, a quarter of it record preparation Flat 10¢/page here, duplicates free Record packet · Case #IME-4812 cited 100% Production342 pp / 2 packets · 11 deduplicatedlogged Disputed requestIndexed to the UR determinationlinked Guideline sectionSupport cited to the MTUS sectioncited Handwritten chart notesRead and transcribed with the restdegraded Formulary entrySurfaced beside the request it coverssurfaced 1 disputed request · cited 100% · no medical-necessity calls What the file actually looks like ### One request. Three hundred pages of context. A single utilization review dispute travels with 50 to 300 pages, 175 in the middle of the range. The dispute itself is narrow: one requested treatment, one denial or modification, one guideline section. Everything else in the packet exists to show whether the record supports it. The pages rarely arrive in one clean production. They come as several: the treating physician's chart, the requesting report, prior authorizations, imaging reports, pharmacy history. The same progress note appears in three of them. Older chart notes are handwritten. Scans of scans lose the dates in the header. In the demo case, 342 pages across 2 packets reduced to 27 documents once 11 duplicate pages came out and page 140, which belonged to a different patient, was quarantined. The volume is not a rounding error. DWC's 2026 IMR report counts 152,351 final determinations in 2025, from 201,037 applications received, and upstream California utilization review decisions run roughly twenty times that. What we'd build with you ### Four steps, starting with one real file. 01 #### Send one real file One disputed request with the productions behind it, as you already have them, in whatever formats they arrived in. Handwriting, faxed scans, and duplicated progress notes included. 02 #### We return the read A sorted, deduplicated packet with the record evidence indexed against the guideline section the request cites, every line traced to its source page. You check it against the file you know. 03 #### We scope the build Your packet template, your index order, your guideline set, your turnaround clock. If your volume or your rules make the build unworkable, we say so instead of shipping something that misses. 04 #### You run it Your clinical operations team runs the packets and your reviewers decide. We never decide medical necessity, grade a request, or predict what a determination will say. Nearby in workers' compensation: [California QME record review](https://medrecords.ai/solutions/california-qme-record-review/) and [impairment rating record review](https://medrecords.ai/solutions/impairment-rating-record-review/) . Audit trail · Case #IME-4812 exportable 08:12Production received · 342 pp / 2 packetssystem 08:3111 pages deduplicatedsystem 08:44p.140 flagged — wrong patientsystem 08:54Packet assembled · 27 documents · cited 100%system 09:20Packet reviewed · citations verifiedreviewer 09:26Packet exported · determination written 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 ### Four capabilities behind every record packet. The utilization review packet is one deliverable of the same platform that reads, sorts, and cites the whole record. [ #### Document Sorting & Packet Builder Several productions sorted into one indexed packet, in your order, with duplicates and wrong-patient pages pulled out before assembly. IN ACTION · 342 pp / 2 packets → 27 documents, one index ](https://medrecords.ai/product/document-sorting-packet-builder/) [ #### Medical Chronology The conservative-care history a guideline section asks about, built as a timeline the reviewer can read in one pass, synced to every source page. IN ACTION · 7 visits improving since 4/02, in date order ](https://medrecords.ai/product/chronology/) [ #### Literature & Standard-of-Care Matching The guideline section and published literature a request points to, put beside the record evidence that speaks to it. The reviewing physician still applies it. IN ACTION · guideline section → the pages that address it ](https://medrecords.ai/product/medical-literature-standard-of-care-matching/) [ #### Verifiable AI Citations If we can't cite it, we don't say it. Every line in the packet index links back to the page range it was pulled from. IN ACTION · 27 documents → cited 100% ](https://medrecords.ai/product/citations/) FAQ ### Utilization review record packets, answered. The disputed treatment request, the utilization review determination that denied or modified it, and the medical records that support the request. Labor Code 4610.5 sets the application and the documents that travel with it. We index the production, remove duplicate pages, and cite each supporting finding to the page it sits on, so the reviewer is not hunting for it. No. We assemble, index, and cite the record. Medical necessity is decided by the utilization review physician and, on dispute, by the independent medical reviewer at Maximus. We never score a request, predict a determination, or make an eligibility call. The MTUS at 8 CCR 9792.20 et seq incorporates the ACOEM Practice Guidelines and the MTUS Drug Formulary, and ODG is the usual reference where the MTUS is silent. We surface the record evidence the cited guideline section asks about, page-cited. Choosing the controlling guideline and applying it stays with the reviewing physician. We do not claim that it does. DWC publishes the number annually: 10.2% of determinations overturned in 2025, down from 12.7% in 2024. What a complete, indexed, page-cited packet does is make sure the reviewer sees what is actually in the file before that determination is written. 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 packet build to minutes. ### Send one 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=utilization-review-record-packet) [Book a demo](https://medrecords.ai/demo/?src=utilization-review-record-packet) [See the whole workers' compensation lane →](https://medrecords.ai/solutions/workers-comp/) [Every case type we process](https://medrecords.ai/solutions/case-types/)