2013 Fiscal Year Final Research Report
Statistical research on design of clinical trials where the exact time to event can not be obseved
| Project/Area Number |
23500365
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Statistical science
|
|---|
| Research Institution | National Institute of Public Health |
|---|
Principal Investigator |
NISHIKAWA Masako 国立保健医療科学院, 研究情報支援研究センター, 上席主任研究官 (50373395)
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
TANGO Toshiro 昭和女子大学 (70124477)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | 医薬生物 / ゲノム統計解析 / 医療 / 臨床試験 |
|---|
| Research Abstract |
As common practice, if the exact time of occurrence is not known due to interval-censoring, a single point imputation (e.g. right-point imputation) for the censoring interval is most often done as though it was the exact time of event and the standard survival analysis are performed. On the other hand, there are several corresponding methods to interval-censored data. Zao's method is one of them. It was found that Zao's method does not maintain type I error probability in our preliminary research. We explored characteristics of single point imputations and Zao's method deeply, and reported the results at an international conference. We tried to improve Zao's method. While using R to evaluate an improved statistics, we found unstable performance of a package (IntCox). We consider some remedies for 'IntCox' and will report the results at an international conference.
|
