AWS for Industries
Enhancing patient experience: AI-driven healthcare scheduling on AWS
In today’s rapidly evolving healthcare landscape, the intersection of patient engagement and scheduling efficiency represents one of the most critical, yet challenging, aspects of healthcare accessibility and overall healthcare delivery. Healthcare organizations across the spectrum—from small practices to large hospital systems—continue to grapple with the dual imperatives of enhancing patient satisfaction while optimizing operational workflows.
According to a recent health survey only 25% of patient scheduling operation have been partially or fully automated, rest actions are still performed manually at patient access centers. These centers are responsible for organizing access to care, bridging clinician’s availability and managing communication with patients. Healthcare call centers handle an average of 2,000 calls daily, with peak staffing levels meeting only 60% of required coverage, resulting in an abandon rate of 7%. That translates into a daily revenue loss of up to $45,000 for providers.
The traditional scheduling process compounds an already stressful situation during a patient’s journey. A patient may make multiple phone calls during work hours to coordinate the appointment post insurance authorization, while also absorbing complex pre-procedure instructions over the phone. Patients are often navigating the uncertainty about next steps, while managing the anxiety of dealing with their diagnostic results.
We will demonstrate how Amazon Connect and Amazon Bedrock AgentCore can provide healthcare providers with a way to automate the scheduling process, resulting in a simplified healthcare journey for the patient. Through the integration of contact centers, electronic health records (EHR), and other existing hospital systems, providers can harness the power of data and agentic AI. This can optimize clinical resources, reduce abandonment rates, and improve overall patient satisfaction.
The automated AI-driven workflow
The patient experience when leveraging an automated workflow is different: when a patient’s prior authorization is approved, Amazon Connect automatically initiates the scheduling workflow. Strands Agents, an open-source, autonomous AI agent framework running on Amazon Bedrock AgentCore, coordinate behind the scenes. They check available appointment slots, confirm procedure requirements, and prepare outreach—all while maintaining HIPAA compliance and patient privacy controls.
Strands Agents streamline the appointment management operations by automatically initiating contact post prior authorization. The solution will:
- Offer flexible scheduling options
- Provide clear procedure information
- Manage follow-up communication
- Facilitate compliance throughout
The automated workflow leverages Amazon Nova Sonic, a speech-to-speech foundation model that handles the conversation in near real-time, speaking naturally with the patient to schedule the appointment.
Two characteristics of Amazon Nova Sonic that are useful for healthcare implementation are:
- Managing human-like conversational flow: Amazon Nova Sonic recognizes conversational subtleties (such as natural pauses, hesitations, and turn-taking cues), allowing it to respond at appropriate moments and seamlessly manage interruptions during the conversation. This capability is particularly important in healthcare settings where patients may need time to process information or formulate questions.
- Implementing knowledge retrieval: For healthcare applications requiring access to specific medical information, Amazon Nova Sonic supports tool use and agentic Retrieval-Augmented Generation (RAG) with Amazon Bedrock Knowledge Bases. This allows voice agents to retrieve accurate medical information during conversations.
The following steps explain how the preceding architecture functions:
- The workflow begins when the EHR database, hosted on AWS HealthLake, triggers an Amazon EventBridge event upon the data entry of a prior authorization verification.
- When Amazon EventBridge identifies a matching data entry event, it automatically triggers an AWS Lambda function.
- The AWS Lambda function makes a call to the Amazon Bedrock AgentCore API, which initiates the Orchestrator Agent’s execution, passing necessary prompts and parameters.
- The Orchestrator Agent processes the prompts and parameters. It acts as a central manager that directs and coordinates a team of other specialized AI agents to complete complex tasks:
- Eligibility Agent
- Appointment Scheduling
- Appointment Reminder
- AgentCore Memory is a fully managed service that enables the agents to retain past interactions for more personalized conversations, handling both short-term context and long-term knowledge retention.
- AgentCore Observability provides comprehensive monitoring, tracing, and debugging capabilities for production environments.
- The agents utilize a set of tools to interact with external systems, enabling them to access critical information (such as appointment schedules, patient preferences, and provider availability).
- The Appointment Scheduling and Appointment Reminder agents trigger outbound calls through Amazon Connect contact flows.
- Amazon Connect outbound call architecture leverages several Amazon Web Services (AWS) services to enable high-volume and efficient outbound communication. Amazon Connect can reach EHR database, hosted on AWS HealthLake, or any other external EHR’s APIs to update patient information or appointments.
Agentic AI architecture deep dive
This architecture utilizes Strands Agents, that can be deployed at scale using Amazon Bedrock AgentCore. Amazon Bedrock AgentCore provides the infrastructure for production-ready agents and handles tasks such as memory, authentication, and observability, so developers can focus on building agents with their preferred framework.
The Orchestrator Agent processes prompts and parameters by interfacing with foundational models, tools, and data sources through the Amazon Bedrock AgentCore Gateway to achieve its objectives. Leveraging the dual capability of AgentCore Memory for short-term conversation context and long-term preference storage, the solution learns and adapts to individual patterns over time. It can automatically suggest convenient appointment slots, while reducing administrative burden.
This transformation creates a streamlined, patient-centric scheduling experience that benefits both healthcare providers and patients by maintaining context across sessions and delivering personalized interactions. The central Orchestrator Agent maintains context across the conversation and verifies requests are handled by the most suitable sub agent.
The Eligibility Agent (4a) validates patient scheduling readiness by analyzing multiple data sources, including EHR demographics, active encounters, and procedure-specific timing rules. Once verified, it seamlessly integrates with operational calendars to check physician availability, facility capacity, and equipment schedules before flagging eligible patients for the appointment scheduling agent to initiate outreach.
The Appointment Scheduling agent (4b) invokes Amazon Connect voice calls to patients identified as ready by the Eligibility Agent. It fetches comprehensive patient data, including contact preferences and available time slots from the EHR database. It manages the entire scheduling workflow, while flagging any unsuccessful contact attempts for follow-up.
The Appointment Reminder agent (4c) operates according to established operational rules, sending timely reminders at specified intervals before appointments, while confirming critical details such as date, time, and location. Each communication is timed to align with provider-specific instructions and facility protocols. Through comprehensive tracking mechanisms, the system monitors reminder delivery status and logs the patient’s responses, so healthcare providers can proactively address potential no-shows or scheduling conflicts. This systematic approach facilitates consistent patient engagement, while maintaining detailed records of all communication attempts and outcomes.
Achievable results
This approach fundamentally reframes the patient scheduling experience from a transactional interaction to a thoughtful touchpoint in the care journey. When patients receive timely, personalized communication that respects their time and preferences, it sets a positive tone for their entire care experience.
The financial implications of our solution are compelling. By reducing appointment abandonment rates and streamlining operations, healthcare providers can recapture significant lost revenue. They can also redeploy staff to higher-value activities that require human empathy and clinical expertise. Moreover, this technology provides healthcare organizations with a way to scale their patient engagement capabilities without proportionally increasing staffing costs, a crucial advantage in today’s challenging healthcare labor market.
As healthcare continues to evolve toward value-based models, these seemingly small improvements in patient engagement become increasingly vital to both clinical outcomes and organizational success.
Conclusion
By automating routine scheduling processes, while maintaining the human touch that patients value, healthcare organizations can simultaneously address operational inefficiencies and enhance the patient’s experience. The integration of Amazon Connect and Amazon Bedrock AgentCore with healthcare providers call centers, EHRs and hospital systems, demonstrates how AI-driven workflows can reduce administrative burden, minimize abandoned calls, and create more meaningful patient interactions at critical moments in their healthcare journey.
The future of healthcare scheduling isn’t just about filling appointment slots—it’s about creating seamless, supportive patient experiences that reduce anxiety and build trust.
Ready to transform your healthcare administrative processes? Explore the AWS Strands Agents SDK documentation and Amazon Connect to discover how AI can revolutionize your organization’s approach to healthcare scheduling. Contact an AWS Representative to find out how we can help you achieve your organizational goals.