A buyer's guide
Risk adjustment coding services: capture the HCCs that determine the payment
In risk adjustment, an uncaptured diagnosis is not a coding error. It is revenue that never existed, on a member you are still responsible for. HCC coding is the one place where what you fail to document costs as much as what you code wrong.
Reviewed by Amanda Chuderewicz, Director of Coding & Auditing Services
Risk adjustment coding is the capture of the diagnoses that drive risk-adjusted payment. Under the CMS-HCC model used for Medicare Advantage and the HHS model used for ACA plans, certain chronic and serious conditions map to Hierarchical Condition Categories (HCCs), and those HCCs set a member's risk adjustment factor (RAF), which determines the plan's payment.
It is different from fee-for-service coding in one way that changes everything. A missed HCC does not trigger a denial. The claim pays, and the payment is simply lower than the member's actual acuity warranted, for the whole year, with no signal that anything went wrong.
Two rules make it hard. Chronic conditions have to be documented and coded every calendar year to keep counting, so last year's capture does not carry forward. And RADV audits can claw back payment for any HCC the documentation does not support. Accuracy has to run both ways: capture what is there, and never code what the chart does not support.
Where risk adjustment leaks
Four ways RAF accuracy is lost
Risk adjustment revenue leaks quietly, because none of these produce a denial that anyone notices.
Uncaptured chronic conditions
A documented chronic condition that never gets coded is an HCC that never counts. The member's RAF understates their acuity, and the payment follows the RAF.
Missed annual recapture
HCCs reset every calendar year. A condition captured last year that is not recaptured this year drops off the RAF, so the revenue has to be earned again on every member, every year.
Unsupported codes and RADV risk
Coding an HCC the documentation does not support is worse than missing one. RADV audits can claw the payment back, so over-capture is a liability, not a cushion.
Specificity gaps
Many diagnoses only map to an HCC at a specific level of documentation. A vague code captures no HCC where a specific one would, and the gap is invisible on the claim.
How it works
How MediCodio codes for risk adjustment
Reads the full record for documented conditions
NLP reviews the encounter and the problem list for chronic and serious conditions that are documented and addressed, including the ones a fee-for-service pass would not need to capture but risk adjustment depends on.
Maps diagnoses to HCCs at the right specificity
Documented conditions are coded to the specificity that captures the correct HCC, and conditions that are named but not supported to the required level are flagged rather than coded, to keep capture defensible.
Checks capture against the member's history
The system surfaces chronic conditions captured in prior years that have not been recaptured this year, so the annual reset does not quietly erode the RAF.
Routes the judgement calls to a coder
Where documentation is ambiguous or an HCC carries RADV exposure, the chart holds in CoPilot for a credentialled coder, with the documentation and the rule behind the call in view.
Capture and defend
Capture everything supported. Code nothing that is not.
Risk adjustment accuracy has to run in both directions, and MediCodio's two lanes are built for it. AutoPilot captures the clearly documented HCCs at volume. CoPilot holds the charts where a condition is named but the documentation may not support the code, so a credentialled coder makes the call rather than the RAF carrying a code a RADV audit would remove.
Every coded HCC carries the documentation passage and the rule that supports it, so capture can be defended in an audit rather than assumed. The routing starts conservative and widens only as accuracy is proven on your own members' charts, which is the only sample that predicts how capture behaves on your book.
Accuracy, defined
98% first-pass, measured on production charts
MediCodio reports 98% first-pass coding accuracy across deployments since 2023, measured at the code level on production charts before any correction, against a reference review by AAPC- and AHIMA-credentialled coders (including CRC holders) applying official guidelines and payer policy to the same documentation. Coverage spans 35+ specialties. The figure that should drive a decision is the one produced on your own members' charts during a pilot, audited by your team.
FAQ
Frequently asked questions
What is risk adjustment coding?
Risk adjustment coding captures the diagnoses that drive risk-adjusted payment. Under the CMS-HCC model for Medicare Advantage and the HHS model for ACA plans, certain chronic and serious conditions map to Hierarchical Condition Categories (HCCs) that set a member's risk adjustment factor (RAF), which determines the plan's payment.
What is HCC coding?
HCC coding is the assignment of ICD-10-CM diagnoses that map to Hierarchical Condition Categories. Each HCC carries a weight that contributes to a member's RAF score. Because a missed HCC lowers payment without producing a denial, HCC coding is about complete, documented capture as much as it is about correctness.
Why is risk adjustment coding different from regular coding?
A missed HCC does not trigger a denial, so the revenue is simply lower than the member's acuity warranted, for the whole year, with no signal. HCCs also reset every calendar year and require annual recapture, and RADV audits can claw back payment for any HCC the documentation does not support. Accuracy has to capture what is documented and never code what is not.
How does MediCodio handle risk adjustment coding?
CODIO reads the full record for documented chronic conditions, maps them to HCCs at the specificity that captures the correct category, surfaces conditions captured in prior years that have not been recaptured, and routes ambiguous or RADV-sensitive charts to a credentialled coder. Every coded HCC carries the documentation and rule that supports it.
Is HCC coding with AI compliant and defensible?
MediCodio is HIPAA compliant, ISO/IEC 27001:2022 certified, and Veradigm Connect certified. Every coded HCC carries the documentation passage and the rule behind it, so capture can be defended in a RADV audit rather than assumed, and conditions that are not supported to the required level are flagged rather than coded.
