InvestorsHub Logo
Followers 90
Posts 17394
Boards Moderated 0
Alias Born 09/06/2006

Re: jammyjames post# 211853

Friday, 02/01/2019 6:17:47 PM

Friday, February 01, 2019 6:17:47 PM

Post# of 699697

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.
Volume:
Day Range:
Bid:
Ask:
Last Trade Time:
Total Trades:
  • 1D
  • 1M
  • 3M
  • 6M
  • 1Y
  • 5Y
Recent NWBO News