Learn More
The trial was designed to detect differences of 10% in parameters evaluated (mean over time) between groups of patients. Type I error (a) was set to 5% and power of statistical tests (1 ) b) to 80%.(More)
  • 1