FAIR Principles in Practice

Data collection is an important aspect of healthcare and is carried out throughout the entire care process. Significant amounts of data are collected and generated throughout this process, potentially relevant for research. However, this data is not immediately suitable for reuse in research. The Findable, Accessible, Interoperable, and Reusable (FAIR) principles, published in 2016, were introduced to describe various aspects of data reuse. Although many steps have been taken to facilitate the implementation of FAIR data, several challenges remain in making data FAIR.
This thesis aims to contribute to the improvement of the implementation of the FAIR data principles in practice. Chapter 2 implements the FAIR data principles in child and adolescent mental health research to identify the most challenging barriers. Chapter 3 assesses the FAIRness of two Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) databases on the European Health Data & Evidence Network (EHDEN) portal, which aims to enhance their findability and accessibility. Chapter 4 evaluates the impact of data element name characteristics on the performance of Usagi, an automatic annotation tool developed by Observational Health Data Sciences and Informatics (OHDSI). Finally, Chapter 5 analyzes two vocabulary data repositories—the OHDSI Standardized Vocabularies and the Unified Medical Language System (UMLS) Metathesaurus— to determine the completeness of their ATC to RxNorm mappings and highlights issues in mappings and annotation exercises.