Opportunities and challenges for biomarker discovery using electronic health record data

Abstract

Electronic health records (EHRs) have become increasingly relied upon as a source for biomedical research. One important research application of EHRs is the identification of biomarkers associated with specific patient states, especially within complex conditions. However, using EHRs for biomarker identification can be challenging because the EHR was not designed with research as the primary focus. Despite this challenge, the EHR offers huge potential for biomarker discovery research to transform our understanding of disease etiology and treatment and generate biological insights informing precision medicine initiatives. This review paper provides an in-depth analysis of how EHR data is currently used for phenotyping and identifying molecular biomarkers, current challenges and limitations, and strategies we can take to mitigate challenges going forward.

Keywords: biomarker discovery; electronic health records; phenotyping; precision medicine.

Figure 1.. Anatomy of the ideal patient electronic health record.

This ideal example of a patient chart contains routinely collected structured data including billing-related data (gray), unstructured notes, oral history, and messaging data (blue), consumer health data and molecular omic data (green), and environmental and SDOH data (purple). Data type-specific methods, validation tests, and interpretations are needed to identify biomarkers effectively. Icons in patient chart customized using BioRender.com.

Figure 2.. Biomarker discovery workflow.

Integration of existing patient data with other health variables can create a more comprehensive picture of patient health. Transforming structured and unstructured data using knowledge sources generates an interpretable clinical context. Lastly, implementation of machine learning and statistical models on large-scale, diverse patient data can enable biomarker discovery. Icons in patient chart customized using BioRender.com.