Ren, Yanjie
(2015)
Interim monitoring and conditional power in clinical trials.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
Large-scale randomized clinical trials are usually used to compare therapeutic effect of a new treatment to that of a standard treatment.
Interim analyses are often performed at several occasions prior to the definitive analysis so that investigators can stop a trial early for ethical or economic reasons
if the accumulating data shows enough evidence of superiority or futility of the new treatment.
It has been recognized that the boundaries for the test statistics at those interim analyses need to be adjusted so that the overall type I error can be properly controlled. In contrary to adjustment for multiple testing in general practice, the theory on adjustment on boundaries for interim analyses have been well developed in past decades because of their sequential nature. At an interim analysis, one may be interested in estimating the chance for demonstrating the expected efficacy benefit from the new treatment at the definitive analysis. Conditional power provides such assessment based on currently available empirical data. Here we review and compare the operating characteristics of some fundamental methods and extensions in regulating the spending of exit probabilities at interim analyses under the null so that the overall type I error is controlled at the desired nominal level. Then we review the development and calculation of conditional power under a few typical settings. We have applied a few methods on planning of interim analyses and the usage of conditional power to two trials from the National Surgical Adjuvant Breast and Bowel projects. Well-planned and scientifically-sound early termination of clinical trials save lives, time and expense in the development of better treatments or regimens. Appropriate use of interim analysis design and conditional power, as we promote here, has tremendous public health significance in improving the lifespan and life quality of the population.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
29 June 2015 |
| Date Type: |
Publication |
| Defense Date: |
23 April 2015 |
| Approval Date: |
29 June 2015 |
| Submission Date: |
6 April 2015 |
| Access Restriction: |
1 year -- Restrict access to University of Pittsburgh for a period of 1 year. |
| Number of Pages: |
42 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Public Health > Biostatistics |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
interim monitoring, conditional power, clinical trail |
| Date Deposited: |
29 Jun 2015 14:37 |
| Last Modified: |
15 Nov 2016 14:28 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/25174 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[img]](http://d-scholarship.pitt.edu/style/images/fileicons/application_pdf.png)
![[feed]](/images/feed-icon-32x32.png){kind=icon}