Head Injury Mortality - Study 0017
The sponsor did an analysis that identified differences in baseline factors between the vehicle and tirilazad groups.
The sponsor asserted that after adjustment for imbalances in baseline characteristics, no difference in mortality was found between the two treatment groups. This conclusion was based on a change in p-value from a significant level to a non-significant level (p=.11).
FDA repetition of this analysis confirmed that adjustment for baseline characteristics (GCS, CT scan findings, pupil reactivity, SBP, age) did raise the p-value above .05
However, FDA analysis also showed that RR only changed 10% (1.37 to 1.23), suggesting that the imbalanced distribution of baseline characteristics across treatment groups did not fully explain the mortality excess in the TIR group