A FHIR Specification to Formalize Cohort Definitions

Area of Science:

• Medical Informatics
• Clinical Research Methodology

Background:

• Retrospective studies are crucial for medical research.
• Cohort definitions are often unstructured, hindering reproducibility and data reuse.
• Lack of standardization limits automated processing of clinical data.

Purpose of the Study:

• To present a framework for structured and standardized specification of cohort definitions.
• To improve reproducibility, consistency, and automated reuse of cohort data.
• To support FHIR (Fast Healthcare Interoperability Resources) based cohort definition.

Main Methods:

• Systematic review of retrospective studies to derive modelling categories.
• Development of a framework using six categories: demographics, terminology, clinical results, temporal data, temporal relationships, and logical combinations.
• Implementation using native or minimally extended FHIR properties.

Main Results:

• A framework enabling structured and standardized cohort definitions within FHIR resources.
• One-to-one correspondence between cohort definitions and clinical data.
• Enabled human readability and automated processing of cohort data.

Conclusions:

• The proposed FHIR-based framework addresses the need for standardized cohort definitions in retrospective studies.
• Facilitates automated processing for use cases like feasibility searches and transparent documentation.
• Enhances the reliability and reusability of clinical data in research.