## 2:1 allocation [General Sta­tis­tics]

Hi Sundar,

welcome back to the forum!

» Will there be any issue in adopting 2:1 (Test:Reference) randomization method in biosimilarity (BE) studies?

You will loose some power compared to equally sized treatment arms. For an -script see there. It gives for an assumed CV 40% and T/R-ratio 0.95 targeted at 80% power:

n = 130 (1:1 allocation)   nT = nR = 65   power = 0.8035 n = 132 (naïve 2:1 allocation)   nT = 88, nR = 44   power = 0.7618 n = 147 (2:1 allocation)   nT = 98, nR = 49   power = 0.8060

If we desire a 2:1 allocation and want to preserve power, we need 13% more subjects than for the 1:1 allocation.

» Note: The 2:1 randomization is mainly due to transition arm.

What do you mean by that?

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes