Statistical methodology for evaluation of survival rate improvement by cancer screening and its application to HCC surveillance

2017 Fiscal Year Final Research Report

• Project/Area Number: 16K15357
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Epidemiology and preventive medicine
• Research Institution: Hiroshima University
• Principal Investigator: Akita Tomoyuki 広島大学, 医歯薬保健学研究科(医), 助教 (80609925)
• Co-Investigator(Kenkyū-buntansha): 田中 純子 広島大学, 医歯薬保健学研究科(医), 教授 (70155266) ; 大久 真幸 広島大学, 医歯薬保健学研究科(医), 助教 (20727250)
• Co-Investigator(Renkei-kenkyūsha): SUGIYAMA Aya 広島大学, 大学院医歯薬保健学研究科, 助教 (50778280)
• Project Period (FY): 2016-04-01 – 2018-03-31
• Keywords: リードタイムバイアス / がん検診の有効性 / 生存時間分析 / 指数モデル / 倍加時間

Outline of Final Research Achievements

As for evaluation criteria for cancer screening, the difference of survival rates between screening-detected group and outpatient-detected group contains the effect of lead-time bias. In this study, we proposed lead-time adjusted methodology. Using doubling time of cancer size, “adjusted follow-up period” which is from the time when “cancer becomes the minimum size to be detected” to time of“death” were used for survival analysis.
We also applied the method to data of HCC surveillance-detected group (N=2,822) and out-patient group (N=1,077), and there is still a significant difference in survival rates after adjusting lead-time. Therefore, HCC surveillance is thought effective to improve the survival rate.

Free Research Field

医学統計学