Improving Data Quality in Oncology Electronic Health Record (EHR)-Derived Databases for Research and Regulatory Decision Making
Author(s)
Zhaohui Su, PhD, Lisa Herms, PhD, Aniketh Talwai, MBA, Jessica Paulus, ScD, Amy K. O'Sullivan, PhD, Sarah Spark, RN, Jennifer Frytak, PhD, Robyn Harrell, MS, Jay Hugalavalli, BS, Janet Espirito, PharmD.
Ontada, Boston, MA, USA.
Ontada, Boston, MA, USA.
OBJECTIVES: The European Medicines Agency (EMA) recently published a real-world data (RWD) quality framework for regulatory decision-making. This study aims to: (1) implement the FDA-led initiative on the oncology quality, characterization, and assessment of RWD (QCARD) to evaluate the relevance, reliability, and external validity of EHR data, and (2) facilitate the use of robust RWD and real-world evidence (RWE) in oncology with the application to other therapeutic areas.
METHODS: The RWD quality dimensions from the FDA-led QCARD initiative were assessed and implemented using Ontada’s ON.Genuity RWD platform. This platform integrates EHR data from approximately 500 US community oncology clinics with external mortality data and allows linkage to claims data at the patient level.
RESULTS: The assessment of relevance revealed consistent high availability of over 250 standardized variables across more than 20 clinical domains for 780,000+ patients over the past 10 years. This data can describe patient demographic and clinical characteristics, treatment patterns, and clinical outcomes. Regarding data reliability, at least 80% completeness was achieved for common variables by enhancing structured data with chart abstraction and natural language processing, with slight variations across cancer types. Conformance was achieved by standardizing ON.Genuity EHR data using Fast Healthcare Interoperability Resources (FHIR) and Minimal Common Oncology Data Elements (mCODE). Mortality data showed high consistency with the external National Death Index data, with a p-value of 0.9 and identical median overall survival for metastatic pancreatic cancer patients.
CONCLUSIONS: The quality and fit-for-use of RWD should be carefully evaluated for each study, including a detailed assessment of relevance, reliability, and external validity in relation to the research objectives. The QCARD initiative represents a significant advancement in EHR-based oncology research, enhancing RWE from RWD and improving decision-making in clinical practice and policy. This research serves as a model for other therapeutic areas and underscores the importance of rigorous data quality standards.
METHODS: The RWD quality dimensions from the FDA-led QCARD initiative were assessed and implemented using Ontada’s ON.Genuity RWD platform. This platform integrates EHR data from approximately 500 US community oncology clinics with external mortality data and allows linkage to claims data at the patient level.
RESULTS: The assessment of relevance revealed consistent high availability of over 250 standardized variables across more than 20 clinical domains for 780,000+ patients over the past 10 years. This data can describe patient demographic and clinical characteristics, treatment patterns, and clinical outcomes. Regarding data reliability, at least 80% completeness was achieved for common variables by enhancing structured data with chart abstraction and natural language processing, with slight variations across cancer types. Conformance was achieved by standardizing ON.Genuity EHR data using Fast Healthcare Interoperability Resources (FHIR) and Minimal Common Oncology Data Elements (mCODE). Mortality data showed high consistency with the external National Death Index data, with a p-value of 0.9 and identical median overall survival for metastatic pancreatic cancer patients.
CONCLUSIONS: The quality and fit-for-use of RWD should be carefully evaluated for each study, including a detailed assessment of relevance, reliability, and external validity in relation to the research objectives. The QCARD initiative represents a significant advancement in EHR-based oncology research, enhancing RWE from RWD and improving decision-making in clinical practice and policy. This research serves as a model for other therapeutic areas and underscores the importance of rigorous data quality standards.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
RWD101
Topic
Health Policy & Regulatory, Health Technology Assessment, Real World Data & Information Systems
Topic Subcategory
Data Protection, Integrity, & Quality Assurance, Reproducibility & Replicability
Disease
Oncology