Publication
A FHIR Specification to Formalize Cohort Definitions
Britta Berens; Joscha Grüger; Carolin Poschen; Konstantin Knorr
In: Studies in Health Technology and Informatics, Vol. 332, Pages 165-169, IOS Press, Niederlande, 10/2025.
Abstract
Retrospective studies play an important role in advancing medical
research, yet especially cohort definitions are often provided in unstructured text
form or in a non-standardized format. This lack of formalization hinders
reproducibility, consistency, and automated reuse. Therefore, we present a
framework for the structured and standardized specification of cohort definitions
within FHIR resources. Drawing from a systematic review of retrospective studies,
we derived six modelling categories for cohort definitions (1) patient demographics,
(2) standardized medical terminology, (3) clinical results definition, (4) temporal
data representation, (5) temporal relationships and dependencies, and (6) logical
combination of criteria. Each category is implemented using native or minimally
extended FHIR properties, establishing a one-to-one correspondence between
cohort definitions and the original clinical data. This enables both human readability
and automated processing, supporting use cases such as feasibility searches and
transparent cohort documentation in study publications.