Power considerations for trials of two experimental arms versus a standard active control or placebo.
Publication
, Journal Article
Hasselblad, V
Published in: Stat Methods Med Res
October 2016
The power of the two-experimental arm trial depends on three choices: (1) when one arm is dropped (if at all); (2) the final testing procedure, assuming no dropping; and (3) the sampling ratio for the three arms. Multiple-arm designs require critical values which were calculated using Mathematica. Power calculations were exact based on probabilities from binomial distributions. The "drop the loser" strategy is optimal for the primary endpoint. The equal sized two treated arm trial gives reasonable power for the primary as well as good power to select the best treated arm. The best power was provided by the 3:3:4 sampling, but it was only marginally better.
Duke Scholars
Published In
Stat Methods Med Res
DOI
EISSN
1477-0334
Publication Date
October 2016
Volume
25
Issue
5
Start / End Page
1824 / 1835
Location
England
Related Subject Headings
- Stents
- Statistics & Probability
- Sample Size
- Probability
- Placebos
- Peptides
- Immunoglobulin Fab Fragments
- Humans
- Eptifibatide
- Controlled Clinical Trials as Topic
Citation
APA
Chicago
ICMJE
MLA
NLM
Hasselblad, V. (2016). Power considerations for trials of two experimental arms versus a standard active control or placebo. Stat Methods Med Res, 25(5), 1824–1835. https://doi.org/10.1177/0962280213503916
Hasselblad, Vic. “Power considerations for trials of two experimental arms versus a standard active control or placebo.” Stat Methods Med Res 25, no. 5 (October 2016): 1824–35. https://doi.org/10.1177/0962280213503916.
Hasselblad V. Power considerations for trials of two experimental arms versus a standard active control or placebo. Stat Methods Med Res. 2016 Oct;25(5):1824–35.
Hasselblad, Vic. “Power considerations for trials of two experimental arms versus a standard active control or placebo.” Stat Methods Med Res, vol. 25, no. 5, Oct. 2016, pp. 1824–35. Pubmed, doi:10.1177/0962280213503916.
Hasselblad V. Power considerations for trials of two experimental arms versus a standard active control or placebo. Stat Methods Med Res. 2016 Oct;25(5):1824–1835.
Published In
Stat Methods Med Res
DOI
EISSN
1477-0334
Publication Date
October 2016
Volume
25
Issue
5
Start / End Page
1824 / 1835
Location
England
Related Subject Headings
- Stents
- Statistics & Probability
- Sample Size
- Probability
- Placebos
- Peptides
- Immunoglobulin Fab Fragments
- Humans
- Eptifibatide
- Controlled Clinical Trials as Topic