Randomization for Clinical Trials

This tool enables the user to choose a randomization procedure based on sound scientific criteria. It comprises the generation of randomization sequences as well the assessment of randomization procedures based on carefully selected criteria. Furthermore, 'randomizeR' provides a function for the comparison of randomization procedures.


The R package randomizeR is an easy to use tool for the randomization of clinical trials. Its aim is to provide sound scientific evidence for the choice of a randomization procedure, on the basis of well-founded criteria. It includes functions for generating randomization sequences from various randomization procedures as well as various issues for the assessment of the randomization procedures.

Installation

You can install randomizeR from CRAN:

install.packages('randomizeR', dependencies = TRUE)

Motivation

Until now, the choice of a randomization procedure did not follow strict scientific principles. If the choice of a randomization procedure does not account for possible problems during the conduct of a clinical trial, the estimation of the treatment effect can be biased. randomizeR models the possible problems that occur in the analysis of a clinical trial, and provides functions to assess randomization procedures according to the issues.

Usage

library(randomizeR)
?randomizeR
vignette("comparison-example")
 
cg <- corGuess("CS")
rar <- genSeq(rarPar(6),10)
assess(rar, cg)

License

The package is free and open source software, licensed under GPL 3.

News

Changes in randomizeR version 1.4

Bug Fixes

  • Minor Bug fixes

New Features

  • Desirability-Functions
  • Selection Bias support for K > 2
  • Chronological Bias support for K > 2
  • Assessment of chronological and selection bias for K > 3

Changes in randomizeR version 1.3

Bug Fixes

  • CreateParam needs parameter N for design PBR.

New Features

  • CreateParam takes parameter filledBlock for Design RPBR and RTBD.
  • Added new vignette for the Assessment and Implementation of Randomization in Clinical Trials (paper submitted to the Journal of Statistical Software)

Minor changes

  • Improved help functions for issues.
  • Improved clarity of vignette comparison-example.

Changes in randomizeR version 1.2

Bug Fixes

  • Efrons Bias Coin Admits odd N

Minor Changes

  • Changed test setting so that tests allocate less memory
  • Added authors of the current version

Changes in randomizeR version 1.1

Bug Fixes

  • Random truncated binomial design produces NA.

New Features

  • Added a reference card containing an overview of the functions of randomizeR, see vignette(randomizeR-refCard ).
  • Added randomization procedures, see ?randPar:
    • Adjustable Biased Coin Design
    • Bayesian Biased Coin Design
    • Generalized Biased Coin Design
    • Chens Design.
  • Included combined additive combination of selection bias and chronological bias, see ?combineBias.
  • Included the function saveAssess() to save the output of the assess function to a file.

Minor Changes

  • Changed show function of assess objects, i.e. appearance of output.
  • saveRand() now saves randomization list as column vector

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("randomizeR")

1.4 by Diane Uschner, 7 months ago


Browse source code at https://github.com/cran/randomizeR


Authors: David Schindler [aut], Diane Uschner [aut, cre], Martin Manolov [ctb], Thi Mui Pham [ctb], Ralf-Dieter Hilgers [aut, ths], Nicole Heussen [aut, ths]


Documentation:   PDF Manual  


GPL (>= 3) license


Depends on methods, ggplot2, plotrix

Suggests testthat, knitr


See at CRAN