Statistical quality control in clinical trials.

We propose to apply statistical methods of industrial process and quality control to accumulating data in a statistical monitoring center of a clinical trial. We discuss some specific issues connected with the application of these methods over calendar time to patients' characteristics (at a particular individual patient time) or to more formal monitoring characteristics (like the number of queries per case record form). The tools used are Shewart charts, plots, breakpoint regression, and recursive residuals with cusums and V-charts, applied to measurement and event data. A software program based on SAS macros allows easy application of the methods. Some examples, with graphical outputs, from an ongoing trial on patients with primary malignant melanoma, illustrate the methods.