Presentation (CTI)

OBJECTIVES: The use of real-world data in Japan has increased substantially since 2010, driven by the availability of large-scale healthcare claims databases for real world evidence. Despite the widespread use of administrative claims databases, database-specific characteristics and methodological limitations may affect study validity. This study aims to systematically characterize oncology research using major Japanese healthcare databases, including the National Database (NDB), Diagnosis Procedure Combination (DPC) database, JMDC, and Medical Data Vision (MDV), with respect to study design, research objectives, and oncology settings.
METHODS: A systematic literature review was conducted in PubMed to identify oncology studies conducted using Japanese healthcare databases (NDB, DPC, JMDC, or MDV) and published in English-language within the previous 5 years. The screening was focused on retrospective observational studies. Primary objective is to evaluate study design, oncology-specific outcomes, and methodological trends. Secondary objective is to compare the strengths and limitations of the databases and evaluate their suitability for specific research applications.
RESULTS: Overall, 382 studies were identified. The interim findings demonstrated distinct patterns in database utilization across oncology research settings. The DPC and hospital-based databases, including MDV, were primarily used for evaluating acute-care outcomes and treatment effectiveness, whereas the NDB was more frequently applied to nationwide epidemiologic analyses. Key limitations included incomplete mortality information in certain databases and limited integration of precision oncology data. The analysis is ongoing and final results will be reported during the presentation.
CONCLUSIONS: Japanese healthcare databases play an increasingly important role in generating real-world evidence in the oncology research. However, robust application requires careful consideration of database-specific strengths and limitations, particularly for advanced methodologies such as machine learning and survival analyses. This research will provide an evidence-based insight to support researchers, clinicians, and pharmaceutical stakeholders in selecting the most appropriate Japanese claims database for the real-world research.