Optimal Spiking Experiment for Noninferiority of Qualitative Microbiological Methods on Accuracy With Multiple Microorganisms

The European and United States Pharmacopoeia demand a noninferiority study on the detection of microorganisms when an alternate qualitative microbiological method is intended to replace compendial method. However, without imposing any modeling assumptions or constraints, studies require large numbers test samples for proposed criterion 0.7 higher each microorganism. When we can assume that accuracy with respect homogeneous across microorganisms, joint statistical analysis data from all be used help reduce sample size dramatically. For this situation, provide statistic noninferiority, optimal spiking experiment, calculation approach under only mild microorganism-specific proportions. We illustrate our real dataset demonstrate good performance using simulation studies.