A surveillance of medical papers to correct inappropriate use of multivariate analysis and plan support for researchers

Research Project

Project/Area Number	26460764
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Research Field	Epidemiology and preventive medicine
Research Institution	The University of Tokyo
Principal Investigator	Nojima Masanori 東京大学, 医科学研究所, 准教授 (00457699)
Project Period (FY)	2014-04-01 – 2017-03-31
Project Status	Completed (Fiscal Year 2016)
Budget Amount *help	¥4,030,000 (Direct Cost: ¥3,100,000、Indirect Cost: ¥930,000) Fiscal Year 2016: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000) Fiscal Year 2015: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000) Fiscal Year 2014: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Keywords	多変量解析 / 疫学 / 臨床疫学 / 医学統計 / 生物統計 / 統計専門家 / ロジスティック回帰 / Cox回帰 / 統計教育
Outline of Final Research Achievements	While "multivariate analysis" is frequently used in medical research, it is currently difficult to say that sufficient understanding has advanced. Therefore, we decided to clarify the actual situation of implementation procedure in medical research paper published in academic journal. In this study, we focused on a method called "use only variables that were significant in univariate analysis" as an improper implementation procedure. The frequency was 8.6% for logistic regression and 9.8% for Cox regression. In addition, it was suggested that the risk of inappropriate methods being used increases when the department in which the first author belongs is not a specialized department such as epidemiology (clinical + other vs. epidemiology / public health department; the odds ratio is 5.6, 95% CI: 2.8 - 15.2).