Payors / risk adjustment
Capture Every Condition, Accurately
Risk scores should say how sick your members are — no more, no less. Ana finds conditions the labs, pharmacy and claims support but this year’s codes miss, and flags codes the record doesn’t support, in the same pass.
Fixed scope: this year’s members and last year’s codes, run inside your cloud.
Chase Codes, or Reflect the Record
- Suspect lists built once a year from claims alone
- Chart reviews assigned by volume, not evidence
- Unsupported codes found by the auditor
- Recapture tracked in spreadsheets by coder
- Suspects found in labs, fills and claims as they land
- Charts opened in order of evidence and impact
- Unsupported codes flagged in the same pass
- Every suspect and decision logged for RADV
Find What the Record Supports
A lab result, a prescription and last year’s code can all point to a condition nobody coded this year. Ana reads them together for every member and names the suspect, with the evidence behind it.
- Labs, pharmacy and claims read together
- Last year’s codes checked for recapture
- Every suspect arrives with its evidence
Is anything missing from member M-27104’s 2026 codes?
Likely CKD stage 3: two eGFR results support it, and it hasn’t been coded this year.
- 2025n18.30 coded · ckd stage 3
- 03/14egfr 48 · lab result
- 05/02egfr 51 · lab result
- 2026no ckd diagnosis on any claim
Open the Charts That Matter First
Coders’ hours are the constraint. Ana ranks suspects by the strength of the evidence and what the condition changes, so the first charts opened are the ones most likely to be documented.
- Suspects ranked by evidence, then impact
- Grouped by provider for one query, not ten
- Unsupported codes queued beside them
Where are the suspected gaps concentrated?
Diabetes with complications leads, then CKD and heart failure.
- diabetes w/ compl. 1,120
- ckd stage 3–5 0.8× diabetes w/ compl. 860
- heart failure 0.5× diabetes w/ compl. 540
- copd 0.4× diabetes w/ compl. 410
Add What’s Supported, Remove What Isn’t
Accuracy cuts both ways. Ana sends supported suspects to coders and provider queries, and sends codes without support to compliance — so the submission reflects the record, and the audit finds nothing new.
- Supported suspects to coders and providers
- Unsupported codes to compliance, with the gap
- Every decision logged for RADV
What goes where from this week’s pass?
500 charts to coders, 214 codes to compliance, every step logged.
- coding Top 500 charts, by evidence and impact
- providers Documentation queries for 38 groups
- compliance 214 codes without support in the record
- finance RAF forecast from this week’s data
Find the Gaps in Your RAF
Bring last year’s submissions and this year’s labs, fills and claims. We join them inside your cloud and hand your coders a ranked list — and your compliance team the codes to check.
- 01A 30-minute call to pick the contracts and the sources
- 02We connect inside your cloud under a BAA — nothing leaves your environment
- 03Your coding and compliance teams review the list with us
Where Risk Adjustment Starts
- 01 Suspect identification
Conditions the labs, fills and claims support but this year’s codes miss, with the evidence attached.
- 02 HCC recapture
Last year’s conditions not yet documented this year, by member and by provider.
- 03 RADV readiness
Codes without support in the record found before an auditor does, and every decision logged.
- 04 Chart review prioritization
The charts most likely to be documented opened first, so coders’ hours go further.
- 05 Provider documentation queries
One query per provider with every open suspect on it, not one per member.
- 06 RAF forecasting
Where scores land if nothing changes, refreshed as data arrives.
No. It finds and ranks the suspects and the unsupported codes; certified coders review the charts and make every coding decision.