dani (version 0.1-1)

test.NI: Non-inferiority test on different scales

Description

A function for testing non-inferiority on either the risk difference, log-risk ratio or arc-sine difference scale.

Usage

test.NI(n0, n1, e0, e1, NIm, sig.level=0.025, scale="RD", print.out=TRUE)

Arguments

e0

Number of events in the control arm.

e1

Number of events in the active arm.

n0

Total sample size of the control arm.

n1

Total sample size of the active arm.

NIm

Non-inferiority margin on the specified scale.

sig.level

One-sided significance level for testing. Default is 0.025, i.e. 2.5%.

scale

The scale on which we define the non-inferiority margin. Can be one of "RD" (Risk difference), "RR" (log-risk ratio) or "AS" (arc-sine difference).

print.out

Logical. If FALSE, no output is printed.

Value

The output is a list, containing the estimate, standard error, cofidence interval (two-sided 2*alpha level), Z statistic and p-value. For the riskdifference scale, the CI is provided both with the standard normal approximation and using newcombe 10 method, as per function ci.pd in package Epi.

Details

This is a function to test non-inferiority of an active treatment against the control within a specific NI margin. The margin can be specified on a number of different scales.

Examples

# NOT RUN {
  
 n0<-1000
n1<-1000
e0<-0.05*n0
e1<-0.05*n1
NImRD=0.05
NImRR=log(2)
NImAS=asin(sqrt(0.1))-asin(sqrt(0.05))
alpha=0.025

test<-test.NI(n0=n0, n1=n1, e0=e0, e1=e1, NIm=NImRD, sig.level=alpha, scale = "RD")
test2<-test.NI(n0=n0, n1=n1, e0=e0, e1=e1, NIm=NImRR, sig.level=alpha, scale = "RR")
test3<-test.NI(n0=n0, n1=n1, e0=e0, e1=e1, NIm=NImAS, sig.level=alpha, scale = "AS")

  
# }