Lovisa Berggren: Statistical work and challenges related to planning, conducting and analysing a clinical trial with cross-over design

Conducting a trial according to Good Clinical practice requires input from a statistician (ICH, E9 http://www.ich.org/ ). The  statistician plays an important role in the study through the different stages of the study: Study planning, study conduction, study analysis and study reporting.

This report will describe the statistical work that was done for study 33. Study 33 is a double blinded, randomized crossover study in healthy volunteers with the primary objective:
“The primary objective for this study will be to compare sedation, one hour after first dose in each period, between qutiapine immediate release formulation (SEROQUEL®) and qutiapine extended release formulation (SEROQUEL XR®) during initial dose escalation.”

The crossover design, where all patients receive both treatment options, allows a within patient comparison. But it also introduces statistical challenges such as period effect and possible interaction between treatment and period. This report describes how these risks were handled in study 33. Furthermore this report includes the results from study 33 as presented in a publication of study 33 (attachment 1).