• FLAGGED
  • DATA
  • LISTS
  • HUMAN

How to audit physician financial disclosures with your AI

Connect your AI to federal CMS records to flag suspicious payment patterns. Once connected, your agent handles the data matching and filtering.

Ask AI about this page

Short answer

How can I use AI to audit physician financial disclosures?

You use your AI to pull specific physician records from the CMS Open Payments dataset and compare them against your own lists. This identifies undisclosed payments or high-risk relationships. You'll get a filtered list of flagged individuals and specific payment details.

The audit outcomes.

Where the findings land.

Your AI processes the raw federal data into these specific categories.

FLAGGED

Conflict alerts

The AI identifies physicians whose federal payment records don't match your internal lists. You get a list of names that require immediate review.

DATA

Payment summaries

The AI pulls specific amounts, dates, and sources for payments from the CMS dataset. You receive a clean breakdown of what was paid and to whom.

LISTS

Physician profiles

The AI aggregates all relevant payment history for a specific doctor. You get a consolidated view of their financial ties to manufacturers.

HUMAN

Review queue

The AI flags the discrepancies, but you decide if a conflict is actionable. You take the final step of investigating the intent behind the payments.

The workflow

What your AI does when the data arrives.

The AI moves from broad dataset searches to specific identity matching.

  1. Identify physicians

    The AI finds the correct physician profiles within the federal database using names or NPI numbers.

    search_physicians

  2. Extract payment details

    The AI pulls the specific financial transaction records associated with the identified doctors.

    query_dataset

  3. Cross-reference records

    The AI compares the CMS data against the physician lists you provide in your chat.

    query_dataset

  4. Generate reports

    The AI organizes the findings into a structured list of flagged conflicts and payment totals.

    query_dataset

Try it

Copy these to start.

Use these prompts to kick off your audit.

Starting points

Replace the bracketed text with the specific names or NPI numbers you need to check.

Open Payments (CMS) Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_PFgVygzxyWx8tBYJnrB993CjxJkqslxBuj7JwmHP/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Open Payments (CMS) capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "open-payments-cms-physician-data-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_PFgVygzxyWx8tBYJnrB993CjxJkqslxBuj7JwmHP/mcp"
    }
  }
}
Copy into chat04

  • Search for all payments made to Dr. Sarah Jenkins in the last year.

  • Find the total amount paid to NPI 1234567890 by pharmaceutical companies.

  • Check if any physicians in my list have received payments exceeding $5,000 from manufacturers.

  • List all physicians in the dataset who received consulting fees from Pfizer in 2023.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Open Payments (CMS Physician Data) once, then ask.

One click connects the dataset to your AI client. Your credentials stay encrypted and secure, and you can start querying immediately.

Connect Open Payments (CMS Physician Data) to your AI

FAQ

How the audit task behaves.

  • 01

    Can the AI change the data in the CMS database?

    No. The AI can only read and query the data. It cannot modify or delete federal records.

  • 02

    Does the AI automatically report conflicts to regulators?

    No. The AI only identifies the data. You are responsible for reviewing the findings and deciding whether to report them.

  • 03

    Can I upload my own list of doctors to check against?

    Yes. You can paste a list of names or NPI numbers into the chat, and the AI will use those to filter the CMS data.

  • 04

    What happens if a physician's name is spelled differently?

    The AI uses the search tool to find matches, but you should always verify the NPI number to ensure accuracy.

  • 05

    Can the AI perform real-time audits?

    Yes. The AI queries the dataset directly, so it works with the most recent information available in the Open Payments records.

  • More questions about Open Payments (CMS Physician Data)? The Connector page answers them. See everything the Open Payments (CMS Physician Data) Connector can do

Connect Open Payments (CMS) to Claude, Cursor, ChatGPT & more