Computer-assisted drug assay interpretation based on Bayesian estimation of individual pharmacokinetics: application to lidocaine.

A microcomputer program for individualized drug level prediction based on Bayesian forecasting is presented. It is written so that the clinician can integrate patient demographics and drug levels to design a new dosage regimen tailored to an individual patient. The program's great flexibility and robustness make it appropriate for realistic clinical settings. A validation with a data set of lidocaine concentrations measured in 18 patients revealed that the program can predict serum lidocaine levels accurately enough to enhance individual patient dosage adjustment within a few hours after a dosage regimen is started.