Real-World Data Quality Framework for Oncology Time to Treatment Discontinuation Use Case: Implementation and...

Area of Science:

• Oncology Outcomes Research
• Real-World Evidence (RWE) in Health Informatics
• Data Quality Assessment Methodologies

Background:

• Real-world evidence (RWE) is vital for observational oncology studies.
• Fragmented health IT systems create challenges for data quality and interoperability.
• Systematic data quality evaluation is essential for reliable oncology outcomes research using real-world data (RWD).

Purpose of the Study:

• Implement real-world time to treatment discontinuation (rwTTD) as a novel use case.
• Utilize the Use Case Specific Relevance and Quality Assessment framework for RWD.
• Assess data quality and relevance for fit-for-purpose RWD in oncology.

Main Methods:

• Mapped operational definition of rwTTD to RWD elements from oncology electronic health records.
• Defined 20 tasks to assess completeness and plausibility of data for SACT use, line of therapy (LOT), death date, and follow-up length.
• Applied Use Case Specific Relevance and Quality Assessment to two oncology databases to estimate rwTTD for advanced head and neck cancer patients.

Main Results:

• Dataset A showed higher relevance for estimating rwTTD (24.96% vs 5.92% receiving target SACT).
• Differences observed in SACT drug terminology, LOT format, and distribution between datasets.
• Dataset B exhibited poorer data completeness, data lag, and mortality data, indicating unsuitability for rwTTD estimation.

Conclusions:

• Fit-for-purpose data quality assessment revealed significant variability between RWD sets.
• Established data quality specifications for rwTTD are adaptable for diverse oncology use cases.
• Emphasizes the need for rigorous data quality checks in real-world oncology research.