Novel method for derivation of within-individual CV from a health screening database

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K19436
• 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: Hiroshima University
• Principal Investigator: Kawano Reo 広島大学, 病院(医), 講師 (00744210)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 総合検診 / 基準範囲 / 個体内変動幅 / RCV / 変化量の有意性評価

Outline of Final Research Achievements

Health check-ups are designed not only to detect medical issues, but also to identify risk factors and illnesses before they start to cause problems. However, it is not easy to judge a significance of change observed within each individual. Therefore, it is essential to have specific criteria for evaluating whether any between-year change observed in each screening test is practically significant or not.
In this study, we tried to estimate level-specific within-individual variation or reference change values for health-screening tests from a large long-term health-screening database. For this purpose, researchers invented an efficient and stable method for estimating within-individual variation.

Free Research Field

臨床検査医学、生物統計学

Academic Significance and Societal Importance of the Research Achievements

本研究で得られた検査値レベル別の個体内変動幅及びそれに基づく基準変化値(RCV)は、検査値変化量の有意性を客観的に判断できる新たな指標である。人間ドック学会の基本検査項目に対する判定区分に応じた形式で成果を取りまとめたため、今後広く利用されることが期待される。
また、本研究は臨床検査値の個体内変動幅の研究の深化に貢献する。具体的には、健診データベースを利活用し、個体内変動幅を検査値レベルに対し連続的に推定する改良法を開発した。先行研究に対し、分割幅可変アルゴリズム及び重み付き推定を取り入れることで、効率的かつ安定した推定と推定範囲の拡大を可能とした。