Opt Vanc is an observational study of critically ill children prescribed IV vancomycin for a suspected infection at the Children's Hospital of Philadelphia. This study will evaluate how well Bayesian dose adaptation, based on a previously-developed population pharmacokinetic (PK) model for vancomycin and a single optimally timed vancomycin concentration, can predict vancomycin area under the curve (AUC) in critically ill children. Eligible subjects will be prescribed vancomycin and undergo routine therapeutic drug monitoring (TDM) per standard of care. At the time of TDM, each subject will have a vancomycin concentration obtained at the most informative sampling time to estimate AUC, as determined by the multiple-model optimal sampling function in PMetrics (population PK modeling program). Investigators will then compare the AUC determined using Bayesian estimation and the subject's optimally timed vancomycin concentration to the AUC determined using Bayesian estimation with all available concentrations (TDM samples plus the optimally timed sample). Investigators will also examine how AUC estimation compares to AUC calculated using standard-of-care methods (ie, log-linear equations). Further, Investigators will evaluate how well the population PK model, along with a subject's measured covariates and the optimally timed PK sample, can predict a subject's future vancomycin AUC.
