AI Medicare Billing Fraud — Upcoding Analysis Playbook

A Medicare Administrative Contractor has flagged a home health agency for billing 94% of visits at the highest complexity level (HCPCS G0179) when the national average for similar agencies is 31%. The agency billed $2.8M in the past 12 months. A whistleblower has filed a qui tam complaint.

When to use this playbook

• Use this playbook when the decision looks like the situation above: A Medicare Administrative Contractor has flagged a home health agency for billing 94% of visits at the highest complexity level (HCPCS G0179) when the national average for similar agencies is 31%.
• It is a fit when you have source files in hand and need a structured, reviewable analysis — not a generic chat answer about "Medicare Billing Fraud — Upcoding Analysis".
• Do not use it as a substitute for licensed, legal, clinical, or authorized official judgment in the domain.

What you'll need

• Agency billing records (12 months, all claims by HCPCS code)
• Patient medical records for a sample of 40 G0179 claims
• National and regional billing benchmarks by HCPCS code
• Staffing roster and clinician credential records
• Whistleblower complaint (redacted)

Attachments: Documents (Documents)

You are a fraud analyst supporting a Medicare billing fraud investigation against a home health agency. I am attaching:

Work only from the attached source files. If a conclusion is not supported, say so.

Produce:
1. Calculate the agency's billing distribution by complexity level and compare to national and regional benchmarks—quantify the upcoding variance in dollars.
2. For the 40 sampled claims, assess whether the medical record documentation supports the billed complexity level using CMS documentation guidelines.
3. Identify whether specific clinicians are associated with the upcoded claims or whether it is uniform across the agency.
4. Calculate the extrapolated overpayment using CMS statistical sampling methodology.
5. Tell me what the False Claims Act exposure is for the agency owner and whether this meets the threshold for criminal referral to DOJ.

Call out where independent models are likely to disagree, and list follow-up documents a reviewer should request.

What to expect

• Billing distribution analysis vs. national benchmark
• Medical record documentation gap list
• Clinician-level upcoding analysis
• Extrapolated overpayment under CMS methodology
• False Claims Act exposure assessment

Review before you act

• Validate this output against source files before relying on it: Calculate the agency's billing distribution by complexity level and compare to national and regional benchmarks—quantify the upcoding variance in dollars.
• Validate this output against source files before relying on it: For the 40 sampled claims, assess whether the medical record documentation supports the billed complexity level using CMS documentation guidelines.
• Validate this output against source files before relying on it: Identify whether specific clinicians are associated with the upcoded claims or whether it is uniform across the agency.
• Validate this output against source files before relying on it: Calculate the extrapolated overpayment using CMS statistical sampling methodology.
• Confirm every cited figure, date, counterparty, or requirement against the attached originals — models compress and can drop a qualifier.
• Treat disagreement between models as a review item, especially on classification, materiality, and recommended next action.
• Do not authorize an operational, clinical, legal, credit, or enforcement action solely because the models agree.