Interim Analysis and Power [Two-Stage / GS Designs]
Hi Emrah,
I assume you wrote in your protocol that your way through the decision tree would be dictated by Cmax and not AUCt? You'll be going into stage 2 before declaring success or failure.
Please help me understand the details. I get on basis of Cmax:
PE=88.31% and CV=27.5% with N1=24, and assuming balance you'll have a power of about 65% (alpha 5%, GMR 95% as per Potvin and not 88.31%). Where did I go wrong here?
❝ I recently performed an interim analysis of a two-stage study with Potvin Method C and here are the initial results (power evaluated with alpha level of 0.05)
❝
❝ LN AUCT: 77.81 - 96.71 Power: 91.55%
❝ LN CMAX: 77.24 - 100.97 Power: 73.51
❝
❝ Given below the flowchart of Potvin method C, I believe this is a fail case where second stage is not required since power of only end-point is below 80%
❝
❝ Do you think is this the right way interpreting the results ?
I assume you wrote in your protocol that your way through the decision tree would be dictated by Cmax and not AUCt? You'll be going into stage 2 before declaring success or failure.
Please help me understand the details. I get on basis of Cmax:
PE=88.31% and CV=27.5% with N1=24, and assuming balance you'll have a power of about 65% (alpha 5%, GMR 95% as per Potvin and not 88.31%). Where did I go wrong here?
—
Pass or fail!
ElMaestro
Pass or fail!
ElMaestro
Complete thread:
- Interim Analysis and Power Emrah Soner Özdeş 2014-12-24 07:38
- Interim Analysis and Power Helmut 2014-12-24 16:22
- Interim Analysis and Power Emrah Soner Özdeş 2014-12-25 08:36
- Interim Analysis and PowerElMaestro 2014-12-25 12:32
- Interim Analysis and Power Emrah Soner Özdeş 2014-12-26 08:55
- R-code & some thoughts Helmut 2014-12-26 17:51
- R-code & some thoughts Emrah Soner Özdeş 2014-12-29 07:49
- R-code & some thoughts Helmut 2014-12-26 17:51
- Interim Analysis and Power Emrah Soner Özdeş 2014-12-26 08:55
- Interim Analysis and Power Helmut 2014-12-24 16:22