1. Raue, Bioinformatics 31(21), 2015↩︎

  2. Kaschek, dMod, 2016 (https://cran.r-project.org/package=dMod)↩︎

Assess impact of covariates on parameters

This function assesses the changes that covariates introduce on the model parameters, relative to a reference individual. Uncertainty in the estimated fixed effect parameters and covariate coefficients is considered. Per model parameter that is changed by covariates, one plot is done. Showing the uncertainty range for the parameter and the impact of the covariates on this parameter. The horizontal lines correspond to the 95% confidence intervals. Plots are done for both estimated and fixed covariate/parameter relationships. Works for all interfaced NLME tools (NONMEM, MONOLIX, NLMIXR). If called without a filename it will return the ggplot objects as a list. If called with filename it will generate the PDF file at the given location. The data used for the plot are returned as attr(,"plotData") in each list element.

covariateEffect_IQRnlmeProject(
  projectPath,
  filename = NULL,
  Nsamples = 1e+05,
  ClinicalRelevanceLow = 0.8,
  ClinicalRelevanceHigh = 1.2,
  SIGDIG = 3,
  CImethod = c("cov", "typcov", "multi"),
  seed = 1234,
  fontsize = 8
)

Arguments

projectPath

Path to an NLME project folder

filename

Filename for export of results. If filename specified, then output written to file PDF file and as a text file with tabular information. If not provided then outputs are only displayed.

Nsamples

How many samples should be taken from the uncertainty distributions. This number should be much larger than the number of individuals in the analysis dataset, so the covariate information in the dataset is well sampled.

ClinicalRelevanceLow

Lower boundary of a grey box around 1

ClinicalRelevanceHigh

Upper boundary of a grey box around 1 This grey box is drawn around the nominal value of 1 and allows a visual feedback of potential clinical relevance.

SIGDIG

Number of significant digits for table output and other values that are printed in some form.

CImethod

Method to determine the confidence interval of the covariate effect. "cov" = consider only uncertainty of the covariate effect parameter (default), "typcov = consider uncertainty of the typical parameter and covariate effect as independent uncertainties. That is an upper bound for the uncertainty of the individual parameter assumed for a certain continuous covariate quantile (5%, 95%) or covariate category. "multi" = consider only uncertainty of the covariate effect parameter, adjusting standard errors to account for the multiple testing problem. See Xu XS, Yuan M, Zhu H, Yang Y, Wang H, Zhou H, Xu J, Zhang L, Pinheiro J. Full covariate modelling approach in population pharmacokinetics: understanding the underlying hypothesis tests and implications of multiplicity. British journal of clinical pharmacology. 2018 Jul;84(7):1525-34.

seed

Seed for random sampling from uncertainty to make the returned plots and tables fully reproducible.

fontsize

Fontsize for the y-axis text and subtitle.

Value

List with ggplot2 objects with each "plotData" attribute, containing the data that were plotted. List also contains the generated table as IQRoutputTable object.

See also

Other IQRnlmeProject: IQRnlmeEst(), IQRnlmeProject(), addCovariateToModelSpec_IQRest(), addPar_modelSpec(), as_IQRnlmeProjectMulti(), as_IQRnlmeProject(), bootstrap_IQRnlmeProject(), compareModels_IQRnlmeProjectMulti(), convertETAINDIVPRED_IQRnlmeProject(), createDataVPC_IQRnlmeProject(), data_IQRest(), dosing_IQRest(), duplicate_IQRnlmeProject(), eigenvalues_IQRnlmeProject(), exportVirtualSubjects_IQRnlmeProject(), getData_IQRnlmeProject(), getETAs_IQRnlmeProject(), getEst_IQRnlmeProject(), getHeader_IQRnlmeProject(), getIndivParameters_IQRnlmeProject(), getIndivPredictions_IQRnlmeProject(), getModel_IQRnlmeProject(), getObsPred_IQRnlmeProject(), getPopParameters_IQRnlmeProject(), getResults_IQRnlmeProjectMulti(), getResults_IQRnlmeProject(), hasrun_IQRnlmeProjectMulti(), hasrun_IQRnlmeProject(), informationContent_IQRnlmeProject(), is_IQRnlmeEst(), is_IQRnlmeProjectMulti(), is_IQRnlmeProject(), is_MONOLIX_IQRnlmeProject(), is_NLMIXR_IQRnlmeProject(), is_NONMEM_IQRnlmeProject(), modelSpec_IQRest(), outlier_IQRnlmeProject(), plot.IQRnlmeProjectMulti(), plot.IQRnlmeProject(), plotConvergence_IQRnlmeProject(), plotETACOV_IQRnlmeProject(), plotETA_IQRnlmeProject(), plotGOF_IQRnlmeProject(), plotINDIVSIM_IQRnlmeProject(), plotINDIV_IQRnlmeProject(), plotVPC_IQRdataVPC(), pred_IQRnlmeProject(), print.IQRnlmeEst(), print.IQRnlmeProjectMulti(), print.IQRnlmeProject(), print_modelSpec(), run_IQRnlmeProjectMulti(), run_IQRnlmeProject(), sample_IQRnlmeProject(), scm_IQRnlmeProject(), summary.IQRnlmeProjectMulti(), summary.IQRnlmeProject(), summaryComments_IQRnlmeProjectMulti(), summaryCorrelations_IQRnlmeProjectMulti(), summaryCovariates_IQRnlmeProjectMulti(), summaryParameters_IQRnlmeProjectMulti(), vpc_IQRnlmeProject()