Clinical study of the Erlanger silver catheter - Data management and biometry

The clinical evaluation of venous catheters for catheter-induced infections must conform to a strict biometric methodology The statistical planning of the study (target population, design, degree of blinding), data management (database design, definition of variables, coding), quality assurance (dat a inspection at several levels) and the biometric evaluation of the Erlange r silver catheter project are described. The three-step data now included: 1) primary data from the hospital, 2) relational database, 3) files accessi ble for statistical evaluation. Two different statistical models were compa red: analyzing the first catheter only of a patient in the analysis (indepe ndent data) and analyzing several catheters from the same patient (dependen t data) by means of the generalized estimating equations (GEE) method. The main result of the study was based on the comparison of both statistical mo dels.