Methinks creep is associated with Cf loading and temperatures in the higher region of bloody hot.
If you ever get to see a slow mo of a blade off test, you will see that the liberated blade does actually take one or more of the less inclined to leave ones. The big deal is the out of balance force caused by the loss of blade mass whilst the hub continues to spin.
I believe GE had to add spoolies to the inlet attachment flange bolts so they didnt liberate the inlet when the blade went south - these crush and absorb the energy of the deforming fan case. Fails the test if that leaves without an invite
