Development of a platform for clinical epidemiological and economic research using large-scale medical and nursing care related databases

2020 Fiscal Year Final Research Report

• Project/Area Number: 17H05077
• Research Category: Grant-in-Aid for Young Scientists (A)
• Allocation Type: Single-year Grants
• Research Field: Epidemiology and preventive medicine
• Research Institution: The University of Tokyo
• Principal Investigator: Matsui Hiroki 東京大学, 大学院医学系研究科(医学部), 助教 (70608794)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Keywords: 大規模医療データベース / NDB / 介護レセプト / 深層学習 / データフュージョン / 臨床疫学

Outline of Final Research Achievements

The secondary use of large-scale medical data is becoming more and more important. The purpose of this study was to conduct epidemiological research by analyzing the NDB. And to analyze the NDB in conjunction with other databases.
In order to conduct research using the NDB, there was a problem that the data structure was complicated. Therefore, in this study, we defined a data structure that is easy for researchers to use, developed our own system for extracting and converting research data from NDB, and conducted epidemiological research. In addition, we developed a method for analyzing the data without linking the databases by using methods such as deep learning, which has been attracting attention in recent years. In addition, we constructed an actual database with ID-connected analysis.

Free Research Field

医療情報・臨床疫学

Academic Significance and Societal Importance of the Research Achievements

本研究により、今まで十分に実施されてこなかったNDBの長期追跡データを用いた疫学研究の実施が可能となった。
さらに、今までは制度上の問題で実施できなかった、データベース間に患者属性が散らばる場合でのデータ解析を、データベース突合を行わずに実施する方法を考案した。加えて、実際にデータベース間を突合したデータの取得にも成功した。今後、今回の研究を基盤として、さらなる疫学研究の実施が可能となった。