| Project/Area Number |
13670331
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Hygiene
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
MURAMATSU Tsukasa Hokkaido Univ., Grad.School of Med., Prof., 医学部, 教授 (10109423)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KANO Katsumi Taukuba Univ., Post-Grad.School of Med., Prof., 大学院・医学研究科, 教授 (10101312)
|
|---|
| Project Period (FY) |
2001 – 2003
|
| Project Status |
Completed (Fiscal Year 2003)
|
| Budget Amount *help |
¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2003: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2002: ¥800,000 (Direct Cost: ¥800,000)
Fiscal Year 2001: ¥1,300,000 (Direct Cost: ¥1,300,000)
|
|---|
| Keywords | Medical Statistics / Epidemiology / Medical Information / Biometry / Non-parametric test / Multiple Logistic Regression Model / Cox Hazards Model / sillubus / The Rancet / ノンパラメトリック検定 / 多重ロジスティック回帰分析 / コッカスハザード回帰モデル / 医学教育カリキュラム / データ処理 / 統計ソフト |
|---|
| Research Abstract |
【Purpose】
We have focused on techniques of statistical analysis used in academic theses in medicine in Japan and abroad aiming at finding how to achieve improved education of Medical Statistics by reviewing wherether appropriate statistical techniques are chosen for statistical analysis in medical theses as well as by examining whether if there are in fact cases of erroneous uses and whether such a phenomenon derives from the educational system including curriculums of medical study together with clarification of qualitative and quantitative differences among analytical techniques.
【Method】
1.We surveyed actual educational curriculums of Medical Statistics in medical training institutions abroad and compared them with those in Japan.
2.We collected theses using Biostatistics among principal medical journals in Japan and abroad for the last ten years, checked if statistical techniques had been properly selected and studied on cases of incorrect uses.
【Conclusion】
1.Our survey of Japanese medi
… More
cal journals indicates that the top ranking three. headed by "(1)test _X2" followed by "(2)t test and (3)correlation analysis", have 70% share in the total use of analytical techniques. Among journals in European languages "(1)test t" ranks at the top, continued to (2)non-parametric-test, (3)ANOVA (4)Comparison of multiple Groups (5)Logistic regression model analysis.
2.Medical Statistics is given a status as an independent educational subject in 20% (2 units, 30 hours) of the cases nationwide. The subject is often given several unit-hours out of Hygiene or Public Health during the third year or, in some other instances, is combined with Epidemiology or Medical Informatics in which analytical methodology is given from the viewpoint of Epidemiology. On the other hand, in the United States and Germany it constitutes an independent academic unit under the name of Biostatistics or Biometry with its own team of educational staff assigned.
3.Incorrect use of statistical techniques and its inferior quality is more conspicuous in journals in Japanese than those in European languages. As the cause is attributed to the inadequate educational system for Statistics in Medical Faculties, it is desired that practical education using proper software at the site of clinical training be strengthened, not to mention of basic education. Less
|
