NorthWest Biotherapeutics Inc (NWBO)

[exwannabe]

Still scratching my head as to how the hell they ended up with 232 test patients and 99 control if they were randomly assigned as les said...

The only thing I could even possibly come up with is that used a randomization alg that was designed for a trial with many less sites (the original design). With only a handful of sites, it could not have gotten as far out of balance even w/o a design to force/nudge it back in balance.

But increase the number of sites, and the chance of a larger imbalance goes up. That is why trials with a large number of sites would have a strategy in place to keep the balance intact.

It it possible that they simply used a bad randomization design for the trial given the final number of sites?

Hodag (on the other board) who is clearly in the industry scoffed at this. And I totally understand his view. But I do not see the numbers this far out w/o something being out of whack.

Note: It was stratified by site and meth status. The site stratification might have been a bad choice with a large number of small sites, geography might have been better.