Artikel
Optimized sample size recalculation in adaptive study designs
Suche in Medline nach
Autoren
-
Carolin Herrmann
- Charité – Universitätsmedizin Berlin, Berlin, Deutschland
-
Geraldine Rauch
- Charité – Universitätsmedizin Berlin, Berlin, Deutschland
| Veröffentlicht: | 27. August 2018 |
|---|
Gliederung
Text
Introduction: A precise sample size calculation is of major importance for a successful and efficient clinical trial. Under- or overpowering trials should be avoided for ethical and economic reasons. As calculation of the “correct” sample size in the planning stage is based on a number of parameter assumptions, which are related to a certain level of uncertainty, an adjustment of the sample size during an ongoing trial is appealing. After recruiting and evaluating a first sequence of patients, updated knowledge on the required parameters is available which can be used to adapt the sample size or to decide on an early stopping.
However, as the observed interim results are also only random realizations of an unknown true distribution, judging and incorporating the information gained at interim is not straightforward. Current recalculation rules (e.g. [1]) often result in highly variable sample sizes, which is inadequate for practical application.
Methods: Optimizing sample size recalculation in adaptive study designs requires two main steps. First, reasonable performance criteria to judge the quality of a sample size recalculation rule must be worked out. In a second step, new ideas for recalculating the sample size at interim upon these new performance criteria can be formulated.
Results: We present a new performance score (based on average conditional power and sample size as well as their standard deviations conditional on entering the second stage) and compare it to the already existing one by Liu et al. [2]. Therefore, we apply the scores to different designs (e.g. the promising design by Mehta and Pocock [3]) and discuss advantages as well as shortcomings. These ideas will yield new insights into potential optimization strategies for new recalculation rules.
Discussion: Although the idea of sample size recalculation based on updated information from the ongoing trial is appealing, much work is required to make these designs feasible and efficient in practice [3]. We summarize and extend new and old considerations required to address this goal.
The authors declare that they have no competing interests.
The authors declare that an ethics committee vote is not required.
References
- 1.
- Bauer P, König F. The reassessment of trial perspectives from interim data – a critical view. Stat Med. 2006;25:23-36.
- 2.
- Liu GF, Zhu GR, Cui L. Evaluating the adaptive performance of flexible sample size designs with treatment difference in an interval. Stat Med. 2008;27:584-96.
- 3.
- Mehta CR, Pocock SJ. Adaptive increase in sample size when interim results are promising: A practical guide with eamples. Stat Med. 2011;30:3267-84.
