Amazon Connect Health FAQs

Amazon Connect Health currently offers the following capabilities:

• Patient verification (GA) – Delivers self-service identity verification for patients, reducing inbound call-handling time for human staff.

• Appointment management (Preview) – Provides a natural conversational experience for patients to self-serve lookup, schedule, or cancel their appointments. Reduces patient caller wait times, enables after-hours scheduling, and relieves burden on human staff.

• Patient insight (Preview) – Surfaces patient visit-specific insights — including patient summaries, health events since the last visit, HCC (Hierarchical Condition Category) recapture from years of longitudinal patient health records. Reduces the time clinicians spend piecing together information before a patient visit.

• Ambient documentation (GA) – Captures patient-clinician conversations in real time and generates clinical documentation and patient-facing after-visit summaries for clinician review. Integrates into existing workflows, eliminating the need to switch between systems. 

• Medical coding (Preview) – Recommends ICD-10, CPT, and E&M codes with reasoning and citations, reducing billing errors and accelerating reimbursement cycles. 

Amazon Connect Health is a purpose-built healthcare service that delivers fully managed agentic AI for patient engagement and point of care workflows. Separately, Amazon Connect is an AI-powered solution for delivering personalized customer experiences through contact centers across industries. For customers who do not want to take on the science, development, safety, EHR integrations, and ongoing maintenance themselves, Amazon Connect Health handles all of that, delivering production-ready agentic capabilities in days, not months.

Amazon Connect Health can work independently from or integrate with Amazon Connect where needed. For example, patient engagement features of Connect Health can integrate natively with Amazon Connect for organizations using it as their contact center. Point of care capabilities of Amazon Connect Health available via SDK do not require the use Amazon Connect.

For patient engagement use cases: You can enable patient engagement capabilities by configuring and customizing AI agents using the Amazon Connect Health application in minutes. You can test them via a sample Amazon Connect contact center workflow automatically created —whether you're adding them to an existing Amazon Connect instance or spinning up a new one.

For point of care use cases: You can embed Amazon Connect Health directly into your application and workflows through our unified Software Development Kits (SDKs).

Yes, Amazon Connect Health is a HIPAA Eligible service. You can find a list of AWS HIPAA Eligible Services here. For more information about AWS, the U.S. Health Insurance Portability and Accountability Act of 1996 (HIPAA), and using AWS services to process, store, and transmit protected health information (PHI), see our HIPAA Compliance Overview.

Amazon Connect Health includes responsible AI safeguards through three key mechanisms:

1. Source evidence linking: Every insight, note, and medical code are linked to the original data sources, such as the patient's medical history. This allows clinicians to easily verify the accuracy of the information.

2. Escalation to human staff: For patient engagement capabilities, we developed specific guardrails and manual escalation paths for each state throughout the agent decision tree. We escalate requests to manual assistance for situations such as medical emergencies, patient-specific requests (e.g., "I want to talk to someone"), and complex or out-of-scope situations. Standard guardrails along with additional customized guardrails are used to block toxicity, harmful content, prompt and system injection, verification bypass, insurance fraud, and certain keyword-triggered content from input and output data. We also provide customization configurations allowing health systems to define their own manual escalation criteria.

3. Rigorous evaluation: Amazon Connect Health undergoes evaluation layers consisting of automated performance evaluations, manual clinical specialist reviews, source evidence mapping, LLM guardrails (e.g., toxicity and harmful content guardrails), and output format validations (e.g., if a medical coding output contains content other than codes or modifiers, the output will be blocked) to ensure accuracy and safety.

Amazon Connect Health includes agentic capabilities that transform how patients interact with your health system:

Patient verification - Verifies patient identity before allowing access to health information or services

Appointment management (Preview) - Handles scheduling, rescheduling, and canceling appointments

These agents conducts natural, contextual conversations with patients powered by the full range of Amazon Connect contact center capabilities . These agentic features are integrated with real-time access to your EHR to ensure patient health info and up-to-date provider, location info, and appointment availability. You can customize agent configurations and capabilities like scheduling, rescheduling, cancelation or verification attributes based on your patient populations, operational requirements, and security standards.

Currently, Amazon Connect Health patient engagement capabilities supports English only. Our capabilities are optimized for natural, conversational interactions with English-speaking patients. We are evaluating additional language support based on customer needs. If your organization requires multilingual capabilities, please contact your AWS Account Team to share your requirements.

Yes, you can associate or link Amazon Connect Health to an existing Amazon Connect Instance using the ARN during setup. If you don't have an existing Connect instance, Amazon Connect Health setup steps in AWS console will create one for you.

We support Epic through a dedicated Amazon Connect Health app available in the Epic App Market. We also provide feature integration to Cerner, MEDITECH, and other EHRs through Amazon Connect Health EHR proxy service and our data integration partners’ connection APIs.

Patient insights automatically surfaces visit-specific insights including patient summaries, health events since the last visit, and HCC (Hierarchical Condition Category) recapture from years of longitudinal patient health records, reducing the time clinicians spend piecing together information before a patient visit and helping them bill correctly as per the patient’s risk profile.

Each summary is organized into five sections:

• Patient and encounter overview: Active conditions, current medications, allergies, surgical history, and family history.
• Since last visit: New labs, specialist notes, medication changes, and recent interventions.
• Trends: Trends over time for key metrics like A1c, blood pressure, weight, and lab values.
• HCC coding analysis: Previously documented conditions relevant to CMS and HHS risk adjustment that are due for recapture.

Patient insights is designed for outpatient clinical settings, particularly primary care, where clinicians see a high volume of patients and need to quickly get up to speed on each patient's history. It replaces the manual process of reviewing dozens of pages of patient records, which typically takes 8 to 12 minutes per patient, with a concise, AI-generated pre-visit summary that is ready before the patient walks in.

For organizations operating under value-based care arrangements, patient insights adds further value by surfacing previously documented conditions that carry HCC risk adjustment weight under CMS and HHS models that are due for recapture, helping providers to accurately assess a patient’s risk profile.

Patient insights brings together data from two source types:

• FHIR data stores — Structured clinical records from FHIR servers like AWS HealthLake.
• Clinical documents — Discharge summaries, consultation notes, and diagnostic reports stored in Amazon S3, or as DocumentReference and Binary resources in FHIR.

This means clinicians get a unified view even when patient data is spread across multiple systems and formats.

Ambient documentation allows providers to focus on their patients during appointments, not on documentation in their EHR. Providers can focus on what really matters — the clinical conversation — with confidence that everything discussed will be converted into a high-quality, structured clinical note, ready for their review and approval before the next patient visit begins. Amazon Connect Health ambient documentation captures patient-clinician conversations in real time, transcribes the audio, identifies speaker roles, and generates structured clinical notes using AI — all without requiring any specialized recording hardware. It works with microphones already built into laptops and mobile devices.

Ambient documentation supports English across 22 specialties

• Allergy Immunology
• Cardiology
• Dermatology
• Endocrinology
• Gastroenterology
• Hematology/Oncology
• Infectious Disease
• Nephrology
• Neurology
• OBGYN
• Oncology
• Ophthalmology
• Orthopedics
• Otolaryngology
• Pain Medicine
• Pediatrics
• Primary Care
• Psychiatry
• Pulmonology
• Rheumatology
• Surgery
• Urology

Amazon Connect Health unified SDK is built for EHR vendors and healthcare application developers which can be used to embed ambient documentation directly into your existing clinical workflows — no rip-and-replace required. Integration follows a straightforward setup path:

• Provision a domain in the Amazon Connect Health console — your isolated environment for managing provider configurations, users, and integrations.
• Create a subscription for each provider, defining their specialty, preferred note template, and output location.
• Stream audio in real time over an HTTP/2 connection using the SDK. The service accepts audio from standard laptop or mobile device microphones. No additional hardware is required.
• Retrieve structured output — clinical notes, evidence mappings linking each generated statement back to the source conversation, and patient-facing after-visit summaries — delivered automatically to your Amazon S3 bucket when the session ends.

Amazon Connect Health integrates with AWS IAM Identity Center for user management and AWS encryption services for data protection, making it compatible with existing enterprise security practices.