sensitivity {BuyseTest}

Sensitivity Analysis for the Choice of the Thresholds

Description

Evaluate a summary statistic (net benefit, win ratio, ...) using GPC along various thresholds of clinical relevance.

Usage

sensitivity(object, ...)

## S4 method for signature 'S4BuyseTest'
sensitivity(
  object,
  threshold,
  statistic = NULL,
  band = FALSE,
  conf.level = NULL,
  null = NULL,
  transformation = NULL,
  alternative = NULL,
  adj.p.value = FALSE,
  trace = TRUE,
  cpus = 1,
  ...
)

Arguments

object

an R object of class S4BuyseTest, i.e., output of BuyseTest

...

argument passsed to the function transformCIBP of the riskRegression package.

threshold

[list] a list containing for each endpoint the thresholds to be considered.

statistic

[character] the statistic summarizing the pairwise comparison: "netBenefit" displays the net benefit, as described in Buyse (2010) and Peron et al. (2016)), "winRatio" displays the win ratio, as described in Wang et al. (2016), "favorable" displays the proportion in favor of the treatment (also called Mann-Whitney parameter), as described in Fay et al. (2018). "unfavorable" displays the proportion in favor of the control. Default value read from BuyseTest.options().

band

[logical] should simulateneous confidence intervals be computed?

conf.level

[numeric] confidence level for the confidence intervals. Default value read from BuyseTest.options().

null

[numeric] right hand side of the null hypothesis (used for the computation of the p-value).

transformation

[logical] should the CI be computed on the logit scale / log scale for the net benefit / win ratio and backtransformed. Otherwise they are computed without any transformation. Default value read from BuyseTest.options(). Not relevant when using permutations or percentile bootstrap.

alternative

[character] the type of alternative hypothesis: "two.sided", "greater", or "less". Default value read from BuyseTest.options().

adj.p.value

[logical] should p-value adjusted for multiple comparisons be computed?

trace

[logical] Should the execution of the function be traced?

cpus

[integer, >0] the number of CPU to use. Default value is 1.

Details

Simulateneous confidence intervals and adjusted p-values are computed using a single-step max-test approach via the function transformCIBP of the riskRegression package.

Value

An S3 object of class S3sensitivity.

Examples

## Not run: 
require(ggplot2)

## simulate data
set.seed(10)
df.data <- simBuyseTest(1e2, n.strata = 2)

## with one endpoint
ff1 <- treatment ~ TTE(eventtime, status = status, threshold = 0.1)
BT1 <- BuyseTest(ff1, data= df.data)
se.BT1 <- sensitivity(BT1, threshold = seq(0,2,0.25), band = TRUE)
plot(se.BT1)

## with two endpoints
ff2 <- update(ff1, .~. + cont(score, threshold = 1))
BT2 <- BuyseTest(ff2, data= df.data)
se.BT2 <- sensitivity(BT2, threshold = list(eventtime = seq(0,2,0.25), score = 0:2),
                      band = TRUE)
plot(se.BT2)
plot(se.BT2, col = NA)

## End(Not run)