HCC risk adjustment coding assigns diagnosis codes that reflect patient complexity and chronic disease burden in Medicare Advantage populations. Done right, it captures the true cost of care and ensures appropriate reimbursement. Done wrong, it leaves money on the table or triggers audit flags. In 2026, Medicare Advantage plans face tighter scrutiny from CMS and RADV audits that target documentation gaps. Most HCC capture errors don't stem from coder mistakes. They start with incomplete physician documentation and weak query processes that fail to clarify clinical intent.
This post identifies the documentation gaps that kill HCC capture, the query techniques that fix them, and the workflow changes that prevent errors before they reach the coding team.
Why documentation gaps create HCC coding errors
HCC codes require specificity that general medical documentation often doesn't provide. A note saying "diabetes" doesn't tell you whether the patient has type 1, type 2, complications, or controlled status. Without that detail, you can't assign the HCC. The coder can't code what the provider didn't document.
CMS updated the HCC model in 2024 (V28), which changed how conditions map to risk scores. Some conditions that used to trigger HCCs no longer do. Others require more specific language. If your documentation habits haven't caught up, you're missing diagnoses that affect payment.
Documentation gaps cluster around 4 areas:
- Chronic condition monitoring statements that don't specify current status or treatment
- Past conditions listed in problem lists without evidence they're still active
- Complication language that's implied but not stated
- Specificity missing for laterality, manifestations, or stage
These gaps don't always matter for fee-for-service coding. They destroy HCC capture.
The compliance risk of undercoding
Undercoding HCCs costs revenue, but it also creates compliance problems. If your charts show clinical activity for diabetic retinopathy but your codes only capture uncomplicated diabetes, auditors see a mismatch. They assume you're either documenting unnecessary care or failing to code accurately. Both scenarios invite scrutiny.
RADV audits sample charts to validate HCC submissions. If a chart doesn't support the submitted HCC, CMS extrapolates the error across your entire population and demands repayment. One documentation gap in 30 sampled charts can trigger a 7-figure recovery demand.
The 5 most common HCC documentation failures in 2026
These documentation patterns appear in nearly every HCC audit we review. They're consistent across specialties and EHR platforms.
1. Chronic kidney disease without stage
Providers document "CKD" or "chronic kidney disease" without specifying stage 3, 4, or 5. Only stage 3 and higher trigger HCCs. Without the stage, coders can't assign the diagnosis. They're forced to query or drop it entirely.
Fix: require stage documentation anytime CKD appears. Most EHRs support auto-text that prompts stage selection when CKD is entered. If the provider doesn't know the stage, they should reference the most recent GFR or creatinine and let the CDI team calculate it.
2. Morbid obesity lacking BMI support
The diagnosis "morbid obesity" needs a BMI of 40 or higher documented in the same encounter. If the chart shows BMI 38 or doesn't list BMI at all, the code doesn't hold up under audit. CMS specifically looks for BMI documentation in RADV reviews.
Fix: auto-populate BMI in every encounter note. Set EHR alerts when BMI crosses 40 so providers know the diagnosis is supportable. If the patient recently lost weight, document that and remove the morbid obesity diagnosis from the active list.
3. Heart failure without specificity
Notes say "CHF" or "heart failure" without distinguishing systolic, diastolic, combined, or whether it's acute or chronic. ICD-10 requires that detail. Without it, you can't assign an HCC-eligible code.
Echocardiogram reports often contain the specificity the provider's note lacks. Coders shouldn't have to hunt through imaging to find it. If the provider ordered the echo, they should reference the findings in their assessment.
Fix: train providers to document heart failure as "chronic systolic heart failure" or "acute on chronic diastolic heart failure." If they don't know which type, query with the echo report attached.
4. Diabetes complications stated in treatment but not in diagnosis
The chart shows diabetic retinopathy exams, neuropathy medications, or nephropathy labs, but the assessment line just says "type 2 diabetes." The clinical picture supports complications. The documentation doesn't.
This happens when specialists manage the complication but the PCP's note doesn't reflect it. The ophthalmologist documents diabetic retinopathy. The PCP copies forward "type 2 diabetes" from last year's note. You lose the HCC.
Fix: cross-reference specialty notes during annual wellness visits. If a patient sees nephrology, ophthalmology, or podiatry, their diabetes should reflect the complication those specialists are treating.
5. Problem list diagnoses with no supporting documentation in the current encounter
The problem list shows "COPD" from 2018. The current visit note doesn't mention respiratory symptoms, spirometry results, or inhaler use. The diagnosis sits there because no one cleaned up the problem list. Under RADV rules, you need evidence the condition is currently monitored or treated.
Fix: audit problem lists quarterly. Remove conditions that haven't been addressed in 12 months unless they're clearly chronic and stable. If a condition is stable, document it as "stable COPD on maintenance therapy" so the record shows ongoing management.
How to write physician queries that capture HCCs without leading providers
A good query clarifies clinical intent. A bad query tells the provider what diagnosis you want. CMS and OIG have both published guidance on compliant queries. The line between clarification and leading is specific.
Compliant queries present clinical indicators and ask the provider to interpret them. They don't suggest a diagnosis.
Example of a leading query: "The patient has a GFR of 28. Can you document CKD stage 4?"
That's leading. You're giving the answer.
Compliant version: "The patient's GFR is 28 ml/min. Please document the stage of chronic kidney disease or clarify if this finding represents acute kidney injury."
You're presenting the clinical fact and asking the provider to interpret it. That's defensible.
Query structure that works for HCC capture
Every query should include 4 elements:
- The clinical indicator from the chart (lab value, imaging finding, medication, consult note)
- The gap in documentation (missing stage, missing specificity, unclear if active)
- Multiple clinically reasonable options, not just the HCC-eligible answer
- An option to state the condition isn't present or is resolved
If you only offer HCC-eligible choices, auditors will call it leading. Always include a non-HCC option.
Example: "The chart shows prescriptions for gabapentin and annual monofilament exams. Please clarify the indication: diabetic peripheral neuropathy, other neuropathy, or pain management unrelated to diabetes."
That query gives the provider room to document what's clinically accurate. It doesn't push them toward the HCC.
When to query and when to let it go
Don't query everything. Query when clinical indicators clearly support a more specific diagnosis but the documentation is vague. If the chart genuinely doesn't support the condition, move on.
High-value query targets: conditions with strong HCC weights like morbid obesity, CHF, CKD stage 3+, diabetes with complications, and vascular disease with complications. Low-value queries: conditions that don't affect risk score or where the clinical evidence is thin.
Track query response rates by provider. If a provider ignores 80% of queries, your process isn't working. Either the queries are bad or the provider needs education.
Workflow changes that prevent HCC errors before coding
Most HCC errors can be stopped upstream with better clinical documentation improvement processes. Coders shouldn't be the first line of defense. CDI specialists should catch gaps during concurrent review.
Concurrent CDI review for Medicare Advantage encounters
Review charts before they close. CDI specialists can query providers while the patient is still fresh in their mind. Post-discharge queries take longer and get lower response rates.
Focus concurrent reviews on annual wellness visits, chronic disease management encounters, and any visit where the problem list shows HCC-eligible conditions. Those encounters drive risk adjustment. Acute visits for minor complaints don't.
EHR templates that prompt HCC specificity
Build smart phrases and templates that require specificity. When a provider types "diabetes," the template should auto-expand to prompt type, control status, and complications.
Some EHRs support decision trees that show HCC impact in real time. If the provider adds "with diabetic nephropathy," the system flags that it affects the risk score. That immediate feedback changes behavior faster than retrospective education.
Problem list governance
Assign someone to own the problem list. In most practices, problem lists accumulate diagnoses no one ever removes. Old resolved conditions sit next to active ones. Coders and auditors can't tell which is which.
Set a rule: if a condition hasn't been addressed in 12 months, it moves to "inactive" unless the provider confirms it's stable and monitored. That keeps the problem list accurate and prevents coding of conditions no longer under treatment.
Annual wellness visits are the best time to scrub the problem list. The provider reviews all active conditions with the patient. Anything not discussed gets flagged for removal or clarification.
How outsourced HCC coding teams reduce capture errors
In-house coding teams know your providers and your EHR, but they also inherit your documentation habits. An external team brings fresh eyes and specialized risk adjustment expertise that catches gaps your internal staff might overlook.
Certified risk adjustment coders review charts specifically for HCC capture. They're trained to spot clinical indicators that support queries. They know which conditions map to V28 HCCs and which don't. That knowledge matters when documentation is borderline.
External teams also bring volume efficiency. If you only code 200 Medicare Advantage charts a month, your coders don't build the pattern recognition that comes from coding 2,000. Specialists who code HCC full-time see edge cases daily. They know when to query and when the documentation won't support it.
Outsourced coding also scales faster than hiring. If you acquire a new Medicare Advantage contract and need to double HCC coding capacity in 60 days, you can't hire and train that fast. An outsourcing partner already has the bench.
What to look for in an HCC coding partner
Not all coding companies understand risk adjustment. Look for teams with certified risk adjustment coders (CRC credentials), experience with RADV audits, and a query process that's compliant with OIG guidance.
Ask about their approach to physician query workflows. If they don't query or they send queries that lead providers toward specific diagnoses, you're buying an audit risk.
Check whether they provide HCC capture reports that show missed opportunities by provider and by condition. That data drives your CDI and provider education programs. Without it, you're coding blind.
Frequently asked questions about HCC risk adjustment coding
What is HCC risk adjustment coding?
HCC risk adjustment coding assigns ICD-10 diagnosis codes that map to Hierarchical Condition Categories, which CMS uses to calculate risk scores for Medicare Advantage patients. Higher risk scores reflect sicker patients and generate higher capitation payments. Accurate HCC coding requires documentation that proves each condition is currently monitored or treated, not just listed in the patient's history.
How often do HCC codes need to be documented?
CMS requires HCC conditions to be documented at least once per calendar year to count toward that year's risk score. Most plans push for documentation during annual wellness visits, but any face-to-face encounter works as long as the provider addresses the condition. If a chronic condition isn't documented in a given year, it won't contribute to that year's risk adjustment even if it was coded the year before.
What's the difference between HCC coding and regular medical coding?
Regular medical coding focuses on billing for services rendered and typically prioritizes the primary reason for the visit. HCC coding captures all chronic conditions and comorbidities that affect the patient's overall health status, even if they aren't the focus of the current visit. HCC coding requires greater specificity and supporting documentation because CMS audits these codes to validate risk scores and prevent overpayment.
Can coders assign HCC codes without physician documentation?
No. Coders can only assign codes supported by physician documentation in the medical record. If clinical indicators suggest a condition but the provider hasn't documented it, the coder must query the provider rather than assume the diagnosis. Coding without documentation support violates compliance standards and won't survive a RADV audit.
What happens if an HCC code is rejected in a RADV audit?
If CMS rejects an HCC code during a RADV audit because the documentation doesn't support it, they extrapolate that error across your entire member population and calculate an overpayment amount. You're required to repay the excess risk adjustment payments tied to that code. High error rates can trigger expanded audits and ongoing oversight. This makes accurate documentation and compliant coding practices critical to financial stability for Medicare Advantage plans.
Get HCC capture right the first time
HCC errors cost more than lost revenue. They create audit exposure, compliance risk, and provider frustration when queries pile up months after the visit. The fix isn't harder coding. It's better documentation workflows, smarter queries, and specialists who know risk adjustment inside out.
If your HCC capture rates are flat while your patient complexity is rising, you're leaving money on the table. If RADV audits keep finding documentation gaps, you need a process that catches them before coding. MedCodex Health offers certified risk adjustment coding teams and CDI support built specifically for Medicare Advantage populations. Talk to us about a chart review pilot that shows exactly where your documentation gaps are and what they're costing you.