One will make better decisions about a drug's development using high quality clinical data from fewer patients than by basing the decision on mediocre data from a large number of patients.

Challenge yourself and others to reduce the number of inclusion criteria, even for Phase 2 trials where you want a homogeneous population of patients.

While a statistical interpretation of clinical data is done by a statistician, the clinical interpretation of clinical data is done by a clinician. Even if a single person is trained and experienced in both fields the two processes are different.

Analyses of clinical data usually refer to statistical analyses. A clinician is important for making suggestions (to a statistician) of which analyses to conduct, both in advance of the trial being completed and after the initial analyses are completed.

Unexpected or strange interpretations should always be carefully assessed, questioned, and reassessed before they are either accepted or challenged.