Situational Awareness for Novel Epidemic Response
0.1.0 - STU Ballot

This page is part of the Situational Awareness for Novel Epidemic Response (v0.1.0: STU 1 Ballot 1) based on FHIR R4. The current version which supercedes this version is 1.0.0. For a full list of available versions, see the Directory of published versions

Architecture and Implementation

This guide envisions a system architecture comprised of small, easily deliverable components that can be used together to support deployment of reporting, analysis, alerting and data aggregation tools to support fast deployment of solutions supporting public health situation awareness.

The technology uses for SANER includes both FHIR Servers and clients, implemented in or along side other Health IT Systems to support the use cases of this IG. The use cases in turn are supported by the definitions of actors that implement various transactions to implement the functionality described in this guide. These transactions exchange data conforming to profiles and using the vocabulary defined in this guide.

Major Components

The major design components described by this guide support the following functions:

  • Reporting - Generates the data to be reported and make it available.
  • Presentation - Analyzes and display data.
  • Adaptation - Adapts between systems
  • FHIR Server - A FHIR Server is an obvious component in a FHIR based implementation guide.
  • External Storage - Stores data for later access.

These functions are described in more detail in the sections below.

Reporting

The reporting component is responsible for reporting utilization data known by an authoritative system in the hospital. There may be multiple authoritative systems for different kinds of reporting data. A reporting component will create FHIR Measurement Reports documenting the measures that are known to the authoritative system. These reports will be collected in some form of centralized storage or transmission solution. That solution could be as simple as a file system, or more complex (e.g., using cloud-based storage), and could involve different protocols, including S3, Azure Blob Storage, SFTP, or FHIR RESTful APIs and a FHIR Server.

In this guide, the Reporting component is implemented by the Measure Source actor.

Presentation

The presentation component selects some subset of available data and presents it to end users providing those users with the ability to interpret and act on the presented information. These components may generate maps showing hot spots based on some aspect of the measured data, charts and annotations of recent trends (e.g., current values plus projections over time), or other visualizations.

In this guide, the Presentation component is implemented by the Measure Consumer actor.

Adapters

Adapters bridge systems lacking certain capabilities with other systems requiring those capabilities, much like a ground adapter does with two pronged electrical outlet.

In this guide, the Adapter component is implemented by the Measure Intermediary actor.

FHIR Server

Several off the shelf FHIR Server components exist which can be stood up in a data center or in the cloud to provide functionality that supports the FHIR Project. HAPI on FHIR is a Java-based FHIR Server that is at the core of the Smile CDR commercial offering. Microsoft offers the Azure API for FHIR. Firely offers Vonk which has both a free community edition and a commercial product. HL7 offers a page filled with Open Source FHIR Server (and client) implementation software.

The figure below shows a FHIR Server acting as an Intermediary between a Hospital and Public Health

FHIR Server

However, a Hospital could simply configure it’s own FHIR Server, and Public Health could then pull the data they need from it.

PublicHealthPullFromHospital

Or, Public Health can configure it’s own FHIR Server, and a Hospital can push the data to it.

HospitalPushToPublicHealth

External Storage

Deployments can also use external storage, for example a file system, message queue, S3 bucket, Azure Blob Storage, SFTP, or even Direct Messaging to exchange data. In the figure below, a hospital pushes data to its own external storage, and public health pulls data from it.

Hospital Pushes To its own External Storage

But a hospital might as easily push data to public health’s external storage system.

Hospital Pushes To Public Health Agency's External Storage

Deployment

System components are designed such that they can be individually deployed and configured.

The following diagrams depict several different deployments with different responsibilties owned by a Hospital or Public Health Agency

Hospital Supported Infrastructure

Hospital Supported Infrastructure

Middle of the Road

Middle of the Road

Public Health Supported Infrastructure

Public Health Supported Infrastructure