DK MedCom Document, published by MedCom. This guide is not an authorized publication; it is the continuous build for version 2.0.0 built by the FHIR (HL7® FHIR® Standard) CI Build. This version is based on the current content of https://github.com/medcomdk/dk-medcom-document/ and changes regularly. See the Directory of published versions

Unstructuredclinicalcontent

Profiling Unstructured Clinical Content in MedCom FHIR Standards

This page provides inspiration and guidance for stakeholders who are exploring the idea of sharing files or other base64-encoded content as part of a MedCom FHIR standard.

The content is intended to support early discussions and considerations around how unstructured clinical information can be exchanged in a consistent and standards-based way within the healthcare sector.

The intention is to illustrate a general and reusable modelling approach. The described pattern can be applied when there is a need to establish a new MedCom standard for sharing document-based or otherwise unstructured information.

When to apply this pattern

This pattern is relevant when there is a need to share clinical content that:

• Is primarily document- or file-based.
• Cannot be represented meaningfully as structured data alone.

The described approach serves as a foundation that can be further constrained and profiled to form a concrete MedCom standard for a specific use case.

Conceptual approach

Unstructured content can be exchanged by using a FHIR Document Bundle as the overall container for the document.

Within this document structure, the MedComDocumentObservation resource can be used to represent the attachment as a clinical artefact. The Observation provides the clinical context, while the actual content is carried using standard FHIR attachment mechanisms.

This approach makes it possible to represent a wide range of attachment types within the same overall document pattern.

What is considered an attachment?

In this context, an attachment may include, but is not limited to:

• Files such as PDF documents or images.
• Base64-encoded content in general.

The examples above are intentionally broad and are meant to illustrate possible types of unstructured.

Why profile documents this way?

Using a Document Bundle combined with MedComDocumentObservation-based profiling allows MedCom to:

• Reuse established and widely supported FHIR patterns
• Support multiple attachment types within a consistent document structure
• Enable fast standardisation when new needs arise

This makes the approach particularly suitable as a starting point when considering new use cases for sharing unstructured data.