Abstract
In this chapter, we consider possible selection bias that can occur due to the fact that a Phase 2 trial is usually selected for use in future trial planning only when it has produced a positive result. We first illustrate the phenomenon of selection bias and relate it to the general concept of regression to the mean. We take a simplified situation where a single Phase 2 trial comparing a single dose versus a control produced a promising treatment effect. We review several approaches that have been proposed to discount the observed treatment effect in the single Phase 2 trial when results from the trial are used to plan a Phase 3 trial. We review the literature that compares these approaches with respect to the remaining bias and the average power of the subsequent Phase 3 trial if a Phase 3 trial is to be conducted. While we have hinted at the existence of selection bias in earlier chapters, Chapter 12 is the first time we consider this issue thoroughly. We offer some recommendations on how to watch out for selection bias when planning for late-stage trials.
…it is the peculiar and perpetual error of the human understanding to be more moved and excited by affirmatives than by negatives.
—Francis Bacon
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Barr, D. R., & Sherrill, E. T. (1999). Mean and variance of truncated normal distributions. The American Statistician, 53(4), 357–361.
Chuang-Stein, C., & Kirby, S. (2014). The shrinking or disappearing observed treatment effect. Pharmaceutical Statistics, 13(5), 277–280.
De Martini, D. (2011). Adapting by calibration the sample size of a phase III trial on the basis of phase II data. Pharmaceutical Statistics, 10, 89–95.
Johnson, N. L., & Kotz, S. (1970). Continuous univariate distributions – 1. Wiley. Chapter 13.
Kirby, S., Burke, J., Chuang-Stein, C., & Sin, C. (2012). Discounting phase 2 results when planning phase 3 clinical trials. Pharmaceutical Statistics, 11(5), 373–385.
Kirby, S., Li, J., & Chuang-Stein, C. (2020). Selection bias for treatments with positive phase 2 results. Pharmaceutical Statistics, 19(5), 679–691.
O’Hagan, A., Stevens, J. W., & Campbell, M. J. (2005). Assurance in clinical trial design. Pharmaceutical Statistics, 4(3), 187–201.
Pereira, T. V., Horwitz, R. I., & Ioannidis, J. P. A. (2012). Empirical evaluation of very large treatment effects of medical intervention. Journal of the American Medical Association, 308(16), 1676–1684.
R Core Team. (2020). A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/
Wang, S. J., Hung, H. M. J., & O’Neill, R. T. (2006). Adapting the sample size planning of a phase III trial based on phase II data. Pharmaceutical Statistics, 5(2), 85–97.
Zhang, J. (2013). Evaluating regression to the mean of treatment effect from phase II to phase III. https://www.amstat.org/meetings/jsm/2013/onlineprogram/AbstractDetails.cfm?abstractid=309685. Accessed 11 Mar 2021.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Chuang-Stein, C., Kirby, S. (2021). Discounting Prior Results to Account for Selection Bias. In: Quantitative Decisions in Drug Development. Springer Series in Pharmaceutical Statistics. Springer, Cham. https://doi.org/10.1007/978-3-030-79731-7_12
Download citation
DOI: https://doi.org/10.1007/978-3-030-79731-7_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-79730-0
Online ISBN: 978-3-030-79731-7
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)