Athenahealth Launches Server Letting US Patients Connect Their Health Records to Claude

The product, called athenaPatient MCP, is a server built on the Model Context Protocol, an open standard that lets AI models pull structured data from authorised systems. Athenahealth said in a Business Wire release that Claude Pro and Max subscribers in the United States can add the connector in their Claude settings from November 2, 2026.

What Claude Can See

Once a patient connects, Claude can draw on lab results, vital signs, medications, health conditions, immunisations, visit history, diagnostic reports, allergies, procedures and care plans to answer questions, according to the company. HIT Consultant, which reported the launch, said the server fetches data on demand when a patient asks a relevant question, instead of loading a full medical history into Claude in advance, and that retrieved information is limited to the active session.

Patients sign in with their athenahealth patient portal credentials, choose which information to share and can revoke Claude's access at any time, the company said. HIT Consultant also reported that Anthropic has zero-data-retention and non-training commitments covering the connection, meaning personal health information is kept out of model training data.

Paul Brient, athenahealth's chief product and operations officer, said in the release: "At its core, this is about helping patients get value from their own healthcare data." Syed Mohiuddin, head of healthcare at Anthropic, said: "Patients often have questions about their health records between visits."

To show demand, the release cites an August 2026 Pew Research Center survey which found that 34% of US adults have used AI assistants for health-related purposes. Neither company said how many athenahealth patients are eligible to use the connector.

Why It Matters

The launch is one of the first cases of a major health record vendor building its own production connection between patient records and a consumer AI assistant, instead of leaving it to third-party developers. Several unofficial athenahealth MCP servers already exist on developer marketplaces, but they are community projects with no vendor backing. Coverage of the launch says Claude is the first assistant athenahealth has connected and that more may follow, though the company has not named any.

For health systems and insurers, the move points to a different way patients will engage with their own data. Questions that used to go to a patient portal message or a phone call to the practice can now be answered by a general-purpose chatbot working from the verified record. That could reduce routine queries for clinicians, but it also raises questions about accuracy, liability and how answers are checked, which the announcement does not address in detail. Some commentators have also pointed out that US patients can already move their records into third-party apps through federally mandated FHIR APIs, so the main change is convenience and verified data inside a mainstream assistant, not new access rights.

For Anthropic, the deal extends its push into healthcare, a market where it competes with OpenAI and others to become the assistant consumers trust with sensitive information. Access is limited to paying subscribers for now, which puts the service behind Claude's paid tiers.