Payment Integrity
Payment IntegrityAsync jobs requiredPre-pay claim review (edits → LLM allegations), policy → edit codification, and COB/MSP primacy.
About
A payment-integrity suite. Pre-pay review screens a claim with deterministic edits (PTP, MUE, duplicate, frequency, modifier, age/sex, bilateral, MOD25, new-patient, missing-diagnosis), then an LLM builds defensible overpayment allegations with supporting and refuting evidence, a defensibility score, and a recommended action.
It also codifies free-text reimbursement policy into an executable, test-validated edit with policy-to-rule traceability, and runs a deterministic COB / MSP primacy determination (birthday rule, working-aged MSP, TPL / subrogation detection).
How it works
- 1/review: deterministic edit screen → LLM allegations with evidence + defensibility (claim JSON or scanned claim + records)
- 2/policy-to-edit: policy text/PDF → executable rule + positive/negative test cases + source-map
- 3/cob-check: member coverage → payer order, primacy, and subrogation flag (synchronous, no LLM)
Intended use
- •Pre-pay overpayment review and recovery routing
- •Codifying payer policy into deployable pre-pay edits
- •Coordination-of-benefits / Medicare-secondary-payer determination
Key outputs
- ▸review → allegations[] (determination, evidence, defensibility, projected_overpayment), claim_determination
- ▸policy-to-edit → edit (when / action), test_results, source_map, deploy_ready
- ▸cob-check → primacy, order[], subrogation_flag, evidence
Endpoints
Try each endpoint with your signed-in session — usage counts toward your monthly budget.
Use synthetic data only. Do not submit real patient records or PHI when testing endpoints.
Limitations & caveats
- –Edit reference values are a curated illustrative subset, not the full CMS tables
- –Allegations are review support, not final payment determinations
- –/review and /policy-to-edit on uploads are LLM-driven (~1 min); /cob-check is instant