Statistical research on design of clinical trial and evaluation reflecting the clinical importance
| Project/Area Number |
18500224
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | National Institute of Public Health |
|---|
Principal Investigator |
NISHIKAWA Masako National Institute of Public Health, Department of Technology Assessment and Biostatistics, Senior Researcher (50373395)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
YANGO Toshiro National Institute of Public Health, Department of Technology Assessment and Biostatistics, Dean (70124477)
|
|---|
| Project Period (FY) |
2006 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥3,940,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥540,000)
Fiscal Year 2007: ¥2,340,000 (Direct Cost: ¥1,800,000、Indirect Cost: ¥540,000)
Fiscal Year 2006: ¥1,600,000 (Direct Cost: ¥1,600,000)
|
|---|
| Keywords | biostatistics / clinical trial / survival analysis / multiple endpoints / test of superiority |
|---|
| Research Abstract |
The aim of this research is to develop a new statistical method for evaluating multiple endpoints (EPs) taking the relative clinical importance between EPs into account. How to measure the relative clinical importance and how to synthesize each component of EPs are important issue.
Investigating the related statistical works from the view point of multiple time to event data and multiple EPs we developed a new composite hypotheses and a composite endpoint. In the case of two primary EPs, a test statistics with respect to the composite endpoint and two test statistics for boundary conditions of respective EPs are formulated. We extended this idea to continuous variables, and new test statistics for them are formulated. Thus far, no attention has been paid to the relative clinical importance of multiple EPs analysis in formulation. We explored a method measuring the relative clinical importance by reading papers and hearing the opinions of medical experts. As an example, we propose a method measuring the relative clinical importance between FEV1 and FVC in pulmonary disease. We evaluate performance of the proposed test by simulation study using software package. The overall size of the proposed test is reasonable in many cases, but it is conservative if true differences are near null hypotheses with respect to at least two test statistics. These results were published at the conferences of International Biometric Society, Joint statistical meeting in Japan and the related symposiums. The result on continuous variables was published at the conferences of East Asia Regional Biometrics Conference as an invited presentation.
|
Report
(3 results)
Research Products
(20 results)
