QualInt (version 1.0.0)

QualInt-package: R-package for qualitative interaction test

Description

Test for qualitative interactions between treatment effects and patient subgroups for continuous, binary and suvival responses. The term treatment effect means a comparison result between two treatments within each patient subgroup.

Details

Package:

QualInt

Type:

Package

Version:

1.0.0

Date:

2014-10-13

This package could be used to calculate the pvalue and power for qualitative interaction testing. Two testing methods are included in the package, which are Interval Based Graphical Approach and Gail Simon Likelihood Ratio Test.

For a complete list of all the functions available in this package with individual help pages, use library(help = "QualInt")

References

Gail and Simon (1985), Testing for qualitative interactions between treatment effects and patient subsets, Biometrics, 41, 361-372.

Pan and Wolfe (1993), Tests for generalized problems of detecting qualitative interaction, Technical Report No. 526, Department of Statistics, The Ohio State University.

Pan and Wolfe (1997), Test for qualitative interaction of clinical significance, Statistics in Medicine, 16, 1645-1652.

Examples

test9 <- qualval(effect = c(1.0, 0.5, -2.0), se = c(0.86, 0.64, 0.32))
print(test9)
plot(test9)

#### Continuous ####
ynorm <- rnorm(300)
trtment <- sample(c(0, 1), 300, prob = c(0.4, 0.6),
                  replace = TRUE)
subgrp <- sample(c(0, 1, 2), 300, prob = c(1/3, 1/3, 1/3),
                 replace = TRUE)
test1 <- qualint(ynorm, trtment, subgrp)
test2 <- qualint(ynorm, trtment, subgrp, test = "LRT")
plot(test1)
print(test1)
coef(test1)
ibga(test1)

#### Binary ####
ybin <- sample(c(0, 1), 300, prob = c(0.3, 0.7),
               replace = TRUE)
test4 <- qualint(ybin, trtment, subgrp, type = "binary")

#### Survival ####
time <- rpois(300, 200)
censor <- sample(c(0, 1), 300, prob = c(0.7, 0.3),
                 replace = TRUE)
test6 <- qualint(Surv(time, censor), trtment, subgrp)