Home/AI Suite/Evaluating HCC Coding AI Vendors
Buyer guide · V28 · MEAT · RADV defensibility

How to evaluate HCC coding AI for risk adjustment. The V28 lens.

Risk adjustment is the highest-stakes coding category in healthcare. RAF scores drive Medicare Advantage payment, ACO REACH benchmarks, and a growing share of Medicaid managed-care revenue. CMS-HCC v28 rolled out across 2024 to 2026 and restructured the condition map. Vendors who are still operating on v24 logic are mis-coding now. Before any pricing conversation, run a V28 currency check. This is the buyer-side guide for everything that comes after.

V28 currency baseline MEAT documentation standard RADV-defensible RAF lift
Qualifying question one

The V28 currency check. Run it first.

CMS-HCC v28 is not a small update. CMS phased it in across payment years 2024, 2025, and 2026. The model retired and consolidated condition categories, the depression and substance-use buckets were restructured, several diabetes complication codes were resequenced, and coefficient weights shifted across the entire map. Vendors who built their HCC engine on v24 and never refreshed it are mis-coding today. They are over-coding conditions that v28 retired and under-coding conditions that v28 added. The downstream effect is twofold. The RAF numbers they report are wrong, and the codes they push into your submissions create RADV exposure when CMS audits.

Before any pricing conversation, ask three questions. What model version are you current on. Do you support a v24 and v28 blended workflow for retrospective coding on older dates of service while running pure v28 for current dates of service. What is your refresh cadence when CMS issues mid-year clarifications and what is the SLA on getting that update into production. A vendor who cannot answer all three precisely is not ready for a Medicare Advantage book.

One more practical filter. Ask the vendor to walk you through ten randomly sampled charts and show you which HCC categories changed between v24 and v28 on those specific charts. The vendors who actually run a current engine can do that in the demo. The vendors who cannot are running marketing slides, not software.

CMS-HCC version timeline · the buyer lens
2023PRIOR YEAR
v24 only
Legacy model. Every HCC vendor was on this. Anyone still here today is selling obsolete code.
2024YEAR ONE
v24 / v28 blend (33% v28)
CMS introduced v28 at one-third weight. Real vendors started running a blended engine.
2025YEAR TWO
v24 / v28 blend (67% v28)
Two-thirds weight on v28. The blended workflow is now table stakes for retrospective work.
2026YEAR THREE
v28 only (100%)
Full transition complete. Current dates of service code to pure v28. Retrospective on 2023 dates still needs v24.
Vendor must support both engines simultaneously through 2027
The buyer scorecard

Ten questions. Ask them in order.

These are the ten qualifying questions that every HCC coding AI vendor must answer with specificity before pricing matters. They are ordered. If the vendor stumbles on question one, the answer to question ten is irrelevant. Print this list. Bring it to every vendor demo.

01

What CMS-HCC version are you current on?

Pure v28 for current dates of service. Blended v24 and v28 for retrospective. Refresh SLA in days, not quarters. Anything less means obsolete coding shipped to your submissions.

02

Do you check for MEAT documentation in every encounter?

The condition must be Monitored, Evaluated, Assessed, or Treated in the current encounter note. Vendors who pull from problem lists without MEAT evidence create RADV risk.

03

What is your RADV defensibility methodology?

Ask for a simulated RADV pass rate on a sample of your own charts. The right metric is audit-defensible RAF lift, not raw lift. Vendors who do not quote a defensibility rate are selling lift you may pay back.

04

Can you run recapture campaigns for prior-year unconfirmed HCCs?

Conditions that were coded last year but not confirmed this year drop off the RAF. The vendor must run recapture as a continuous workflow with clinician-confirmation surfaces, not as an annual report.

05

How do you handle chronic vs acute distinction?

Many conditions code as one or the other depending on documentation language. The AI must surface the linguistic cues correctly and not roll an acute condition into a chronic HCC. Test it on diabetes complications and CHF.

06

What is your false-positive rate?

Over-coding is a RADV nightmare and a clinician-trust killer. Ask for the precision metric (percentage of AI-suggested HCCs that the coder accepts after review). Below 80 percent means too many false positives.

07

Do you support prospective and retrospective workflows?

Prospective surfaces suspected HCCs to the clinician inside the encounter so the documentation is born compliant. Retrospective catches missed HCCs after the encounter closes. You need both, not one.

08

How does it integrate with the EHR problem list?

The AI must reconcile its suggested HCCs against the patient's active problem list, the prior encounter notes, and the current encounter document. Problem-list integration is required, not optional. Ask for the supported EHRs by name.

09

What is your audit trail when CMS comes asking?

For every coded HCC, the system must produce the MEAT sentence, the source document, the timestamp, the coder who confirmed, and the version of the engine that produced the suggestion. Anything less fails RADV.

10

Who owns the coded data and RAF lift attribution?

Read the contract. Some vendors retain rights to the encoded outputs. Some bill on lift attribution they cannot independently verify. You need contractual data ownership and a transparent attribution methodology.

The documentation standard

MEAT. The four pillars of RADV defensibility.

CMS requires that an HCC condition be Monitored, Evaluated, Assessed, or Treated in the encounter note for the code to be RADV defensible. AI that surfaces a condition from a problem list without MEAT evidence is creating audit risk, not revenue. Real HCC coding AI walks the chart and surfaces the MEAT sentence for every coded HCC.

The MEAT standard is the line between defensible coding and audit risk. A condition listed on the problem list without any sentence in the current encounter note that monitors, evaluates, assesses, or treats it does not survive a RADV pull. The auditor will demand documentation contemporaneous with the date of service. If the documentation does not exist, CMS recoups the payment with interest, and pattern-of-conduct violations cascade across the entire plan.

The vendor evaluation question is whether the AI presents the MEAT evidence sentence inline with every suggested HCC. Not whether the AI claims to check MEAT. Not whether MEAT is in the vendor's marketing deck. Whether, on a sample of your own charts, the AI surfaces the actual sentence from the actual encounter document that proves the condition meets one of the four pillars. The vendors who do this are running real natural-language processing against the clinical text. The vendors who do not are pulling from problem lists and calling it AI.

One useful test. Pick a chart where a condition is on the problem list but is not mentioned anywhere in the current encounter note. Does the vendor suggest the HCC anyway? If yes, that vendor is going to fail your next RADV audit.

M

Monitored

Status checked. Ordered labs, reviewed vitals, tracked symptoms. The note shows ongoing surveillance of the condition.

E

Evaluated

Reviewed for change. Considered against history, prior results, current findings. The note shows clinical judgment was applied.

A

Assessed

Documented in the assessment. Carries a status (stable, worsening, improved). The note shows the condition was clinically reasoned about.

T

Treated

Active intervention. Medication adjusted, referral placed, procedure planned. The note shows the clinician acted on the condition.

Audit defensibility

RADV. The only RAF lift that matters.

CMS Risk Adjustment Data Validation audits sample charts every payment year and demand contemporaneous documentation for every HCC submitted. The defensibility metric is the percentage of submitted HCCs that survive the simulated audit. Vendors who quote raw RAF lift without a defensibility rate are selling lift you may have to pay back with interest. Vendors who cannot produce a RADV simulation on a sample of your own charts are not ready for an MA contract.

The right buyer-side conversation does not start at RAF lift. It starts at defensibility rate. Ask the vendor to show you a RADV simulation pass rate on a stratified sample of your charts before signing. A reputable vendor will run the simulation, surface the HCCs that would fail audit, explain why each one failed (typically missing MEAT, problem-list-only sourcing, or specificity mismatch), and quote the defensible RAF lift after the failing HCCs are pulled out.

That is the number you negotiate against. Defensible RAF lift, not raw RAF lift. The difference between those two numbers is the recoupment risk you would otherwise carry on the plan books.

RADV defensibility checklist
01
Contemporaneous documentation. The MEAT evidence must be in the encounter note for the date of service. Backdating fails.
02
Specificity match. The diagnosis code must match the documentation. E11.22 requires renal manifestation in the note. E11.9 (unspecified) does not earn the HCC weight.
03
Credentialed signer. The note must be signed by a clinician credentialed to diagnose the condition. Scribe-only notes without provider attestation fail.
04
No problem-list-only coding. A condition on the problem list without current-encounter MEAT does not survive. This is the most common audit failure.
05
Audit-trail completeness. Source document, timestamp, coder confirmation, engine version. CMS reviewers expect a clean trail per HCC.
The annual cycle

Recapture campaigns. Not an annual report.

Prior-year HCCs that are not confirmed in the current calendar year drop off the RAF score. CMS does not carry chronic conditions forward. Every condition must be re-documented every year, with MEAT evidence, in a fresh encounter, on a fresh date of service. The plans that win on RAF score are the plans that run a continuous recapture campaign, not the plans that run an annual sweep in November.

Real HCC AI surfaces the prior-year unconfirmed HCCs to the clinician inside the workflow. Before the encounter starts, the AI presents a list of conditions that need re-documentation, with the suggested MEAT prompts. After the encounter, the AI verifies that each prior-year HCC was either re-documented with MEAT, ruled out clinically, or flagged for follow-up. This is the workflow that separates real HCC coding AI from a report-running vendor.

The vendor question is whether the recapture workflow runs continuously or whether it dumps a list of unconfirmed conditions on the coder once a year. If the answer is the latter, the vendor is selling a spreadsheet, not software.

Continuous recapture workflow
01
Pre-encounter surface.

Prior-year unconfirmed HCCs presented to the clinician before the visit starts, with suggested MEAT prompts and current chronicity evidence.

02
In-encounter capture.

As the clinician documents, the AI flags MEAT-compliant language for prior-year conditions and prompts when a condition is mentioned without sufficient documentation.

03
Post-encounter verify.

For every prior-year HCC, the AI confirms one of three outcomes: re-documented with MEAT, ruled out clinically with documentation, or flagged for follow-up at the next visit.

04
Quarterly reconciliation.

The recapture rate, the rule-out rate, and the still-open rate roll up to a panel-level dashboard. Open conditions get re-queued for the next encounter automatically.

Full disclosure

We are a vendor too. Honest framing.

This page is a buyer-side guide and we are also an HCC coding AI vendor. We built this guide because we wanted prospective buyers to be able to compare any vendor, including us, against a real specification. The ten questions, the V28 currency check, the MEAT framework, the RADV defensibility rate, and the recapture workflow are the same evaluation criteria we hold ourselves to. If you run the guide against us, we expect you to find some things you like and some things you want to improve. That is the right buyer posture.

Our HCC coding engine runs pure v28 with v24 blend support for retrospective dates. It surfaces MEAT evidence inline with every suggested HCC. The RADV simulation runs on a sample of your charts before contracting. The recapture workflow is continuous, not annual. We integrate with the major ambulatory EHRs and we publish the precision and recall numbers per condition family. Our product page is linked below. Read it after you have read this guide, not before.

Common questions

Frequently asked questions: HCC coding AI vendor evaluation.

What is CMS-HCC v28 and why does it matter for vendor selection?
CMS-HCC v28 is the current risk adjustment model that CMS phased in across payment years 2024, 2025, and 2026. It restructured the condition map, retired some HCCs (notably parts of the depression and substance-use categories were consolidated), introduced new categories, and changed coefficients across the board. A vendor still operating on v24 logic is mis-coding now. The first qualifying question for any HCC coding AI vendor is version coverage: pure v28 for current dates of service, blended v24 and v28 for retrospective work on older dates, and a documented refresh cadence for when CMS issues mid-year updates.
What is MEAT documentation and why do AI vendors need to surface it?
MEAT is the CMS documentation standard for risk-adjustment coding. Every HCC condition must be Monitored, Evaluated, Assessed, or Treated in the encounter note for the code to be RADV defensible. A vendor that surfaces a condition from a problem list or a prior note without checking for MEAT evidence in the current encounter is creating audit risk, not revenue. Real HCC coding AI walks the chart, finds the sentence in the current encounter where the condition is monitored, evaluated, assessed, or treated, and presents that sentence as evidence for the coder.
How is RADV defensibility actually measured?
CMS Risk Adjustment Data Validation audits sample charts each year and demand contemporaneous documentation for every HCC submitted. The defensibility metric is the percentage of submitted HCCs that survive a simulated RADV audit. A vendor that quotes a raw RAF lift number without a defensibility rate is selling lift you may have to pay back, with interest. Ask vendors to demonstrate RADV simulation outputs on a sample of your own charts, with the MEAT sentence cited for every HCC.
What ROI is realistic on HCC coding AI for a Medicare Advantage population?
For a Medicare Advantage population, realistic year-one RAF lift from a well-implemented HCC AI program sits in the 0.08 to 0.18 range on the average member, assuming the baseline operation already has decent prospective and retrospective workflows. Translated to revenue, that is roughly 8 to 18 percent of the average MA capitation rate per member, net of the cost of clinician confirmation time. Vendors quoting 0.30 or higher lifts are typically counting non-defensible codes that will not survive RADV.
Should I focus on prospective or retrospective HCC workflows first?
Both, but in sequence. Retrospective sweeps catch the prior-year unconfirmed HCCs and any missed conditions from completed encounters, so they produce the fastest visible RAF lift. Prospective workflows surface suspected HCCs to the clinician inside the encounter so the documentation is born compliant, which is the durable answer. A vendor that only handles retrospective is incomplete. A vendor that only handles prospective is leaving prior-year revenue on the table. Both workflows must be in scope before signing.
How do I prove RAF lift attribution to the AI vendor?
You need a baseline RAF before the vendor turned on, segmented by member panel, age band, and chronic-condition cohort. You need post-implementation RAF on the same segments. You need a control panel that did not get the AI overlay during the same period. The vendor should provide the segment-level attribution by code group, by clinician, and by panel. If the vendor cannot give you attribution at that grain, the RAF lift number they quote is unfalsifiable and you should treat it as marketing rather than measurement.
Does HCC coding AI matter outside Medicare Advantage?
Yes. CMS uses HCC-style risk adjustment for ACO REACH, MSSP, the Medicare Shared Savings Program, parts of the ACA marketplace via HHS-HCC, and a growing set of Medicaid managed-care programs. Even traditional fee-for-service Medicare uses HCC for quality program scoring and certain bundled payment models. The vendor question for non-MA contracts is which risk model is used, what the documentation requirements are, and whether the AI supports that specific model rather than only CMS-HCC v28.
How does HCC coding AI connect to clinical documentation improvement (CDI)?
Tightly. The same MEAT documentation standard that makes an HCC RADV defensible also feeds the CDI program. A condition documented with specificity (the diabetes coded as E11.22 with renal manifestation rather than unspecified E11.9) carries the HCC weight. HCC coding AI should integrate with the CDI workflow so a single clinician query covers both the inpatient DRG and the outpatient HCC. Vendors who run HCC and CDI as separate workflows force duplicate queries to the clinician and erode trust. Look for a unified clinical-documentation surface.

Get the buyer scorecard. Free.

The full evaluation worksheet with the V28 currency check, the 10 questions, the MEAT framework test, the RADV simulation request template, and the recapture workflow checklist. Bring it to every HCC coding AI vendor demo. A senior partner on the call to walk through scoring with you.