Bayesian Forcasting Model – 2020 Presidential Elections in the United States of America

The article describes and explains a prior centric Bayesian forecasting model for the 2020 US elections.The model is based on the The Economist forecasting project, but strongly differs from it. From the technical point of view, it uses R and Stan programming and Stan software. The article’s focus is on theoretical decisions made in the process of constructing the model and outcomes. It describes why Bayesian models are used and how they are used to predict US presidential elections.

External Evaluation of Vancomycin Population Pharmacokinetic Models at Two Clinical Centers

Background: Numerous vancomycin population pharmacokinetic models in neonates have been published; however, their predictive performances remain unknown. This study aims to evaluate their external predictability and explore the factors that might affect model performance.Methods: Published population pharmacokinetic models in neonates were identified from the literature and evaluated using datasets from two clinical centers, including 171 neonates with a total of 319 measurements of vancomycin levels. Predictive performance was assessed by prediction- and simulation-based diagnostics and Bayesian forecasting. Furthermore, the effect of model structure and a number of identified covariates was also investigated.Results: Eighteen published pharmacokinetic models of vancomycin were identified after a systematic literature search. Using prediction-based diagnostics, no model had a median prediction error of ≤ ± 15%, a median absolute prediction error of ≤30%, and a percentage of prediction error that fell within ±30% of >50%. A simulation-based visual predictive check of most models showed there were large deviations between observations and simulations. After Bayesian forecasting with one or two prior observations, the predicted performance improved significantly. Weight, age, and serum creatinine were identified as the most important covariates. Moreover, employing a maturation model based on weight and age as well as nonlinear model to incorporate serum creatinine level significantly improved predictive performance.Conclusion: The predictability of the pharmacokinetic models for vancomycin is closely related to the approach used for modeling covariates. Bayesian forecasting can significantly improve the predictive performance of models.

Bayesian forecasting for intravenous tobramycin dosing in adults with Cystic Fibrosis using one versus two serum concentrations in a dosing interval

BackgroundIntravenous tobramycin requires therapeutic drug monitoring (TDM) to ensure safety and efficacy when used for prolonged treatment, as in infective exacerbations of Cystic Fibrosis (CF). The 24 hour area under the concentration time curve (AUC24) is widely used to guide dosing, however there remains variability in practice around methods for its estimation.ObjectivesTo determine the potential for a sparse sampling strategy using a single post-infusion tobramycin concentration and Bayesian forecasting, to assess the AUC24 in routine practice.MethodsAdults with CF receiving once daily tobramycin had paired concentrations measured 2 hours (c1) and 6 hours (c2) following end of infusion as routine monitoring. We estimated AUC24 exposures using Tucuxi, a Bayesian forecasting application incorporating a validated population pharmacokinetic model. We performed simulations to estimate AUC24 using the full dataset using c1 and c2, compared to estimates using depleted datasets (c1 or c2 only), with and without concentration data from earlier in the course. We assessed agreement between each simulation condition and the reference graphically, and numerically using median difference (Δ) AUC24, and (relative) root mean square error (rRMSE) as measures of bias and accuracy respectively.Results55 patients contributed 512 concentrations from 95 tobramycin courses and 256 TDM episodes. Single concentration methods performed well, with median ΔAUC24 <2 mg.h.l-1 and rRMSE of <15% for sequential c1 and c2 conditions.ConclusionsBayesian forecasting, using single post-infusion concentrations taken 2-6 hours following tobramycin administration can adequately estimate true exposure in this patient group and are suitable for routine TDM practice.Key Points-In stable adult patients with Cystic fibrosis without significant renal impairment, Bayesian forecasting allows accurate estimation of tobramycin AUC24 using a single blood sample taken 2-6 hours post-infusion with acceptable accuracy, especially when including prior measured concentrations.-A single sample approach with Bayesian forecasting is logistically less complicated than a two-sample approach, and could facilitate best-practice TDM in the outpatient setting.-A more intensive sampling strategy with Bayesian forecasting using two tobramycin concentrations in a dosing interval should be considered in unstable patients, or where observed concentrations deviate significantly from model predictions.