The warfarin data frame has 251 rows and 8 columns of data containing
data on the pharmacokinetics of warfarin, an anticoagulant drug used in the
prevention of thrombosis and thromboembolism.
warfarinThis data frame contains the following columns:
an integer identifying the subject on whom the observation was made
time since drug administration when the sample was drawn (hr)
total dose received by the subject (mg)
warfarin concentration in the sample (mg/L)
observation type (1 for all observations)
weight of the subject (kg)
subject gender (0=female, 1=male)
age of the subject (yr)
The dataset is the PK part of a larger dataset including both warfarin concentrations and prothrombin complex activity (PCA), which measures the decreased coagulation activity resulting from the inhibition of vitamin K recycling, the mechanism of action of warfarin. It contains the concentrations measured in 32 healthy subjects after a single oral dose of warfarin sodium (1.5 mg/kg of body weight). The subjects in the study were sampled at different times over a period of up to 120 hours.
The data is distributed with the Monolix software as a demo for PK/PD modelling. The data has been slightly reformated for R, removing the line at time=0 and filling the amt column with the dose for each subject, following the output of simulx which was used to simulate data from two alternative models to fit this dataset.
Two datasets containing simulated data are associated with the warfarin data. For
each dataset, 1000 simulations of the original data were performed for the computation of npde.
The package contains only the simulated data simwarfarinCov because of size constraints. simwarfarinBase can be downloaded from
the github for npde3.0: https://github.com/ecomets/npde30/blob/main/keep/data/simwarfarinBase.tab
the data in this dataset was simulated according to a base model without covariates: the PK model was a two-compartment model, with first-order absorption and a time-delay. Interindividual variability was modelled as log-normal distributions for parameters Tlag, ka, Cl and V1, and the error model was a combined error model. The parameters were estimated by Monolix.
the data in this dataset was simulated according to a model including several covariates: an age (centered on 30 yr) effect on Cl, a weight (centered on 70 kg) effect on Cl and V1, and a gender effect on V1. The covariate model was built in Monolix.
PDF documentation for npde 3.0: https://github.com/ecomets/npde30/blob/main/userguide_npde_3.0.pdf
data(warfarin)
#Plotting the warfarin PK data
plot(dv~time,data=warfarin,xlab="Time after dose (hr)",
ylab="Warfarin concentration (mg/L)")
Run the code above in your browser using DataLab