In recent years, multiple frameworks for assessing the quality of real-world data (RWD) have been introduced, identifying key dimensions of quality that are crucial for ensuring the accuracy and reliability of RWD. However, there is a lack of understanding of how these frameworks can be applied to large-scale electronic health record (EHR) datasets in practical terms.
The authors summarize critical dimensions of data quality, like accuracy, completeness, and relevance, identified by key regulatory and policy bodies. They then describe how these dimensions are addressed in a large-scale electronic health record (EHR)-based oncology RWD source.
Why this matters
This study tackles the standard dimensions of reliability and relevance in a practical manner, taking into account the need for robustness, scalability, and feasibility. By doing so, the resulting framework can be applied flexibly to other sources of RWD, promoting transparency in assessing the fitness of RWD for specific uses and standardizing the language for implementing data quality measures.