2010 Fiscal Year Final Research Report
Statistical Methodologies for Improving Predictability in Clinical Research and Development of Drugs
| Project/Area Number |
20500255
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
HAMASAKI Toshimitsu Osaka University, 医学系研究科, 准教授 (40379243)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SUGIMOTO Tomoyuki 大阪大学, 医学系研究科, 助教 (70324829)
SOZU Takashi 京都大学, 医学研究科, 准教授 (80408723)
UESAKA Hiroyuki 大阪大学, 臨床医工学融合研究教育センター, 特任教授 (60446250)
|
|---|
| Project Period (FY) |
2008 – 2010
|
| Keywords | 医薬品開発 / 生物統計学 / 標本サイズ / 臨床試験 / 多重評価項目 / 統計的シミュレーション |
|---|
| Research Abstract |
Statistical methodologies for Improving predictability in clinical research and development of drugs with shorten the time, saving the cost, and improving productivity were considered. (1) the exponential cure classification and regression tree (CART) was newly developed to indentify the patients group with higher risks of remissions and recurrence of diseases, especially with application of major depressive disorder and the statistical simulation for assessment of clinical trial designs were performed with the models suggested model. (2) Performances of randomization methods frequently used in clinical trials were theoretically and numerically assessed. In addition, issues in groups sequential design with sample size modification based on multiple co-primary endpoints were discussed. (3) Power and sample size determinations in clinical trials with multiple correlated continuous, binary and time to events endpoints were discussed and some guidance was provided to how to consider the correlation between the endpoints in power and sample size determinations.
|
