Cox proportional hazard analysis with non-uniform samples; power analysis

#1
We have a study involving 10000 patients, 5000 of them treated with drug A and 5000 with drug B. We want to know if drug A is more effective than B.

The median time to event (death) after treatment with drug A is 1000 days and for drug B it is 2000 days.

There are two sampling scenario:

In scenario 1, we are given data (time to event) for 100 patients who received drug A and 100 patients who received drug B. How could we use this data to tell if drug A and B have different effects ? answer: use Cox proportional hazard analysis

In scenario 2, we are given data for 100 patients who died between 800-1200 days after treatment (some of them received A, some B, we pick randomly) and data for 100 patients who died between 1800-2200 days after treatment. How could we use this data to tell if drug A and B have different effects?

In which of the above scenarios we have better power to detect a significant difference between drug A and B ?
 
Last edited: