The 2026 RADV Audit-Readiness Checklist: V28, MEAT, and Evidence on Every Code

RADV (Risk Adjustment Data Validation) audits used to feel like a remote possibility. In 2026 they are a planning assumption. CMS has signaled aggressive expansion of audit volume, and the full phase-in of the CMS-HCC V28 model has reshuffled which conditions carry weight — and which now require tighter documentation to survive review.

The organizations that sleep well are not the ones who code the most. They are the ones who can produce, on demand, the exact clinical evidence behind every submitted diagnosis. Here is the checklist we walk our clients through.

1. Can you trace every code back to a source document?

This is the single most important question. For each HCC you submit, you should be able to answer — in seconds, not days — which document, which page, and which sentence supports it.

If your answer involves a coder re-opening charts and searching, you have an exposure problem. The goal is a permanent, queryable link between the submitted claim → the accepted diagnosis → the verbatim source evidence. That linkage is the backbone of our Audit & Transparency Module.

2. Does every diagnosis satisfy MEAT?

Auditors do not reward a code that merely appears in a problem list. They look for MEAT — Monitoring, Evaluation, Assessment, or Treatment — documented in the encounter. Before submission, confirm each condition shows at least one MEAT element in the note:

• Monitor: symptoms, disease progression, lab trends.
• Evaluate: test results, response to treatment, exam findings.
• Assess: an explicit assessment or clinical status.
• Treat: medications, referrals, procedures, or a documented plan.

The most common failure we see is "copy-forward" problem lists with no current-year MEAT. A 412-character assessment cloned from last visit is an audit liability, even when it is clinically accurate.

3. Are you re-capturing chronic conditions every year?

Risk scores reset annually. A diagnosis documented and submitted in 2024 earns nothing in 2026 unless it is re-documented this year. RADV exposure runs both ways: codes you over-claimed and revenue you left on the table. A disciplined HCC recapture workflow closes both gaps.

4. Can you defend the reasoning, not just the keyword?

"Black box" suggestions are an audit risk in themselves. If a code was surfaced by software, you need to show why — the clinical rationale and the evidence behind it — to a provider, a coder, and a payer alike. We built MedChartScan around explainable, no-black-box AI for exactly this reason: every recommendation carries a plain-English rationale plus the highlighted source.

5. Can your team produce an audit package in one click?

When the letter arrives, the clock starts. Manual chart-chasing across faxes, PDFs, and disparate EHRs is where teams lose nights and weekends. Audit-ready operations generate a complete documentation package — claim, codes, evidence, dates — for the requested members on demand.

6. Is your evidence trail provider- and payer-facing?

The strongest audit posture is collaborative. Giving health plans secure, read-only visibility into the evidence behind your claims reduces disputes and pre-empts record requests. Audit defense should not be a fire drill; it should be a standing capability.

Build audit-readiness into the workflow, not after it

The teams that pass RADV without drama treat documentation integrity as a prospective discipline — caught at the point of care and at coder review — not a retrospective scramble. Every accepted code in MedChartScan is born with its evidence attached, so the audit package essentially writes itself.

*Worried about your RADV exposure under V28? Schedule a live demo and we'll show you how MedChartScan links every HCC to its source evidence — and generates audit packages on demand._