Development of a probability calculation formula to identify the disease name from the electronic medical record information

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K18954
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 58030:Hygiene and public health-related: excluding laboratory approach
• Research Institution: Hirosaki University
• Principal Investigator: Tanaka Rina 弘前大学, 医学研究科, 助教 (40709713)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: がん登録

Outline of Final Research Achievements

In Japan, since 2011, the Ministry of Health, Labour and Welfare has been providing electronic medical record information extracted from the National Database (NDB) based on the "Guidelines for the Provision of Receipt Information and Specific Health Checkup Information." While the electronic medical record information obtained contains disease name, these names often do not accurately reflect the patient's medical history. Therefore, despite receiving electronic medical record information from NDB, conducting accurate analysis based on specific disease names is currently more challenging. This study aims to develop a probability calculation formula to identify disease names from the provided electronic medical record information by comparing it with various disease registration data and elucidating the characteristics of electronic medical record information.

Free Research Field

がん疫学

Academic Significance and Societal Importance of the Research Achievements

電子レセプト情報には傷病名は含まれているものの、その傷病名はカルテ上の病名を正確に反映していないことが多く、例えばがん患者の罹患数をがん登録データと比較した報告では、レセプト情報上のがんは実際の患者数の最大2.9倍多かった。本研究により構築した確率計算式は、疾病登録以外のデータへの応用が可能であることから、さらに予測精度を向上させることで将来的にレセプト情報による解析が容易となり、ナショナルデータベース活用による研究の可能性が広がることが期待される。