Recapture the conditions already in the chart. Then prove them.
Most lost risk-adjustment revenue is not undiagnosed disease. It is chronic conditions documented somewhere in the record that never made it onto a claim this year. The fix is not more diagnoses. It is disciplined recapture, gated on MEAT-supported documentation so every recaptured code survives a CMS RADV look-back.
A condition can be true, documented, and still worth zero
Risk adjustment resets every calendar year. A diagnosis captured last year does not carry forward on its own. If it is not coded from a qualifying encounter this year, the model treats the member as if the condition is gone.
In the record
Prior-year HCC, a problem-list entry, a med that implies the condition, or a specialist note. Real, but not on a 2026 claim.
Flagged, not coded
The engine surfaces it as a recapture candidate. This is a lead, not a code. Nothing bills yet.
MEAT-confirmed
A 2026 face-to-face note shows Monitor / Evaluate / Assess / Treat. Now, and only now, it flows to the claim.
MEAT is the standard that keeps recapture honest
MEAT is the documentation framework auditors read against Section IV of the ICD-10-CM Official Guidelines, which say to code conditions that require or affect patient care, treatment, or management. A condition that is only listed, with no clinical action, fails. Our engine will not recapture it.
Track it
Signs, symptoms, disease progression, labs trending. The note shows the condition is being watched.
Review it
Test results, medication response, exam findings tied to the condition are reviewed and recorded.
Judge it
Clinical assessment, ordering, discussion, counseling, or a status statement in the provider's own words.
Act on it
Medications, therapies, referrals, or a plan. Evidence the condition drove a care decision this year.
Size the MEAT-confirmed opportunity in your panel
Move the sliders to model your own panel. Outputs are illustrative and use your inputs only. Actual value depends on each member's HCC coefficients under the CMS-HCC V28 model and your county benchmark. No figures below are quoted as ASP results.
What survives an audit, and what gets clawed back
The CMS-HCC RADV Final Rule (CMS-4185-F2) removed the fee-for-service adjuster and allows CMS to extrapolate audit findings across the contract. That turns a handful of unsupported codes into a large, projected recovery. The recapture that holds up looks nothing like the one that does not.
| Test | Fragile recapture | MEAT-gated recapture |
|---|---|---|
| Source of the diagnosis | Prior-year claim carried forward | Coded from a 2026 face-to-face encounter |
| Documentation | Problem-list entry only | Monitor / Evaluate / Assess / Treat in the note |
| ICD-10-CM guideline fit | Listed, not managed | Affects care, treatment, or management (Section IV) |
| Specificity | Unspecified default | Highest documented specificity, laterality preserved |
| RADV posture | Extrapolated finding risk | Chart pulls to the same conclusion as the code |
| Provider attestation | Inferred from a medication | Assessment stated in the provider's own words |
The V28 phase-in makes clean recapture the whole game
V28 phase-in begins
CMS starts the three-year transition to the 2024 CMS-HCC model, blending V24 and V28 payment.
Blend deepens
More weight shifts to V28. Constrained coefficients and remapped HCCs mean prior-year capture no longer predicts current-year value.
V28 fully in effect
Per the CY2026 Medicare Advantage and Part D Rate Announcement, risk scores run entirely on V28. Every recapture must be earned in-year on this year's model.
FY2026 ICD-10-CM guidelines live
The dates of service you are recapturing for 2026 are coded under the FY2026 Official Guidelines for Coding and Reporting. The engine codes to the release that matches the DOS, never a stale one.
How the recapture actually runs
A suspect list is easy. A suspect list that only bills what the chart can defend is the product. Every candidate passes through the same gate before it touches a claim.
Surface suspects
Cross-read prior-year HCCs, problem lists, meds, labs, and specialist notes against this year's coded claims to flag the gaps.
Map to V28
Translate each ICD-10-CM code to its CY2026 CMS-HCC V28 category so you know the real coefficient, not a V24 assumption.
Run the MEAT gate
Scan the in-year encounter for Monitor, Evaluate, Assess, Treat evidence. No MEAT means the suspect stays a suspect.
Check specificity
Enforce highest-specificity coding and laterality against the FY2026 guidelines. Never downgrade to unspecified to force a match.
Build the RADV packet
Bind the code to the exact chart page and MEAT language, so a later audit pull lands on the same conclusion.
Release to claim
Only confirmed, documented, specificity-checked recaptures flow. The rest go back to the provider as a documentation query.
The guidelines this model is built on
CMS-HCC Model V28, CY2026 Rate Announcement
The 2024 CMS-HCC risk adjustment model, fully phased in for calendar year 2026 per the CMS Medicare Advantage and Part D Rate Announcement. Governs which conditions map to a payment HCC and at what coefficient.
ICD-10-CM Official Guidelines for Coding and Reporting, FY2026
Effective for dates of service on and after October 1, 2025. Section IV directs coding of conditions that require or affect patient care, treatment, or management, and enforces highest-specificity code assignment.
MEAT documentation standard (Monitor, Evaluate, Assess, Treat)
The industry-accepted framework for demonstrating that a documented condition was clinically active in the encounter, used to judge whether a diagnosis is codeable and audit-supportable.
CMS-HCC RADV Final Rule (CMS-4185-F2)
The Risk Adjustment Data Validation rule permitting extrapolation of audit findings without a fee-for-service adjuster, which is why every recapture must be tied to MEAT-supported chart evidence.
Recapture what is real. Leave what is not.
ASP-RCM's HCC Engine V28 finds the documented-but-uncoded conditions hiding in your record and only bills the ones a RADV auditor would agree with. Cleaner RAF, fewer clawbacks, a chart that backs every code. Let us run a suspect-and-MEAT read on a sample of your panel and show you the defensible number.
Book an HCC recapture review → Talk to a risk-adjustment leadRelated reading
How Policy 8F Went From Draft to De Facto Enforcement in 82 Days
North Carolina's draft Clinical Coverage Policy 8F rewrites Medicaid autism treatment as RB-BHT. See the dated
Read →BriefingA RADV auditor never reads your intent. They read your chart.
A RADV auditor does not read your intent. They read your chart. Here is how the V28 risk-adjustment engine fla
Read →Case studyThe codes moved. The weight moved with them.
A side-by-side comparison matrix of how diabetes and vascular conditions remap from CMS-HCC V24 to V28, so you
Read →