Problem List Is Not a Diagnosis: CHF, COPD, and CKD at the AWV

Assessed and supported now:

• Type 2 diabetes - assessed with a current value (A1c 8.1) and treated (metformin continued). Monitoring + Treatment met.
• Hypertension - assessed (BP 138/80) and treated. Met.

Not supported on this note alone:

• CHF - appears only in the problem list and PMH. A PMH entry describes history, not current status. However, the exam documents bibasilar crackles and 2+ edema, which are clinical indicators that the condition may be active. That combination - clinical indicators without an assessment - is the textbook indication for a query, not for silent submission and not for silent exclusion.
• COPD - PMH and problem list only. No exam finding, no medication change, no monitoring documented at this encounter. Cannot be submitted from this note. Check other encounters in the year before writing it off - the condition may be well supported elsewhere.
• CKD 3b - same problem, plus a data question: was an eGFR resulted and reviewed? If a lab exists and the provider acknowledged it, that supports monitoring. If not, it is PMH only.

Answer to the MA: a problem list entry is a reminder, not documentation of an encounter. Recapturing a condition requires evidence it was evaluated or treated this year - problem list inheritance alone is exactly what RADV audits deny.

Compliant query (single condition, clinically framed):

Patient: [name], DOS 9/18. Documentation supports the following clinical indicators: bibasilar crackles on exam, 2+ pedal edema, and a documented hospitalization for CHF in 3/2024. The assessment does not address CHF status. Please clarify the status of the patient's CHF at this encounter:

• Acute exacerbation
• Chronic, stable
• History of, not currently active
• Unable to determine
• Other: ______

Why it is defensible:

• Cites the specific clinical indicators from this record - crackles, edema, the March admission.
• Asks one condition per query.
• Offers a genuine range of options, including a 'not active' option and 'unable to determine,' so it is not a push toward a code.
• Ties the request to the provider's own documentation, not to the problem list alone.
• Contains no mention of HCCs, risk scores, reimbursement, or payer impact.

What would make it leading and unacceptable:

• 'CHF is an HCC - please document it so we capture the risk.'
• Offering only 'acute' and 'chronic' with no option that allows the provider to say the condition is not active.
• Pre-filling the answer.
• Bundling CHF, COPD, and CKD into one query so the provider cannot answer one independently of the others.
• Verbal queries with no written record; if a verbal query is unavoidable, document it and have it attested.

Reframe the request. The provider is describing a real problem - under-capture if these conditions are in fact active - but proposing an invalid solution. Address both.

Exposure of coding from the problem list:

• Overpayment and False Claims risk. Submitting diagnoses with no encounter-level support is unsupported risk-adjustment capture. RADV and commercial chart audits reverse these, and repeated findings escalate to extrapolated recovery.
• Asymmetric harm. The same practice also loses legitimate revenue when an active condition is documented poorly in the note but never queried.
• Two payers are already watching. The 4% risk score decline plus external flags make 'code the list' the single worst available response.

What to offer instead:

1. Run a query campaign by condition, not by chart - the top three conditions on this provider's panel, one query each, with clinical indicators pulled from the record.
2. Fix the template: add an assessment prompt for each active chronic condition so it is addressed at the point of care.
3. Pre-visit chart prep so the provider sees the candidate conditions before the patient arrives.
4. Show the provider the clinical consequence: crackles and edema with no CHF assessment in the plan is a care gap, not just a coding gap.

Do not: submit from the problem list, or promise that accurate documentation will raise risk scores - that framing converts an accuracy program into an upcoding program.