February, 2003
© 2000, 2001, 2003 Scott S. Emerson, M.D., Ph.D.
Refining the primary endpoint (cont.)
Time to death (censored continuous data)
·
·Would
have heavy early censoring due to logistical constraints of
trauma centers
·
·Might place emphasis on clinically meaningless
improvements in very short term survival
-e.g., can detect differences in 1 day survival
even if no difference at 10 days