smartDTR {smartDesign}

Dynamic Treatment Regimen (DTR) Trial design clinical trial calculations

Description

Dynamic Treatment Regimen (DTR) Trial design clinical trial calculations

Usage

smartDTR(mu_Barm=cbind(G1=c(30,25), G0=c(20,20)),
                     sigsq_Barm=cbind(G1=c(100,100), G0=c(100,100)),
                     nsubject=500,  Barm=c(1,3), type="continuous",  
                     sens=seq(0.5,1, by=0.1), spec=seq(0.5, 1, by=0.1),
                     pG_A1 = 0.8, pG_A2 = 0.8, pran_A1 = 0.5, 
                     pran_Barm = c(0.5, 0.5))

Arguments

mu_Barm

matrix of two named vectors of the means for the two B arms (columns) for the smart DTR trial, with rows as 'G1' and 'G0'

sigsq_Barm

matrix of two named vectors of the variances (sigma-squared) for the two Blevels (columns) for the smart DTR trial, with rows as 'G1' and 'G0'

nsubject

total sample size for the trial

Barm

for the second phase of the trial, the 'B' levels for which the DTR means/variances apply

type

trial response variable type; only continuous is implemented currently

sens

range of sensitivity for smart SST calculations; (0,1]

spec

range of specificity for smart SST calculations; (0,1]

pG_A1

probability of response to therapy given assignment to A1

pG_A2

probability of response to therapy given assignment to A2

pran_A1

probability of random assignment to A1

pran_Barm

probability of assignment to Barms

Details

see details in the reference

Value

An object of the smartDTR S3 class, with the following elements:

dtrdat:

data.frame with sens, spec, mu, sigsq and sample size (n)

sst1:

smartSST object from the first Barm

sst2:

smartSST object from the second Barm

true_mumix:

true mu mixture

true_sigmix:

true sigma mixture

mu_Barm, sigsq_Barm, Barm:

input B-arm, mu, and sigsq for DTR

Author(s)

Jun (Jessie) He, Aberaham Eyman-Casey, Jason P. Sinnwell, Mayo Clinic

References

Jun He, Donna K. McClish & Roy T. Sabo (2021) Evaluating Misclassification Effects on Single Sequential Treatment in Sequential Multiple Assignment Randomized Trial (SMART) Designs, Statistics in Biopharmaceutical Research, DOI: 10.1080/19466315.2021.1883472

Examples

mumat13 <- cbind(G1=c(30,35), G0=c(20,28))
varmat13 <- cbind(G1=c(100,100),G0=c(100,100))

dtr13 <- smartDTR(mu_Barm=mumat13, sigsq_Barm=varmat13,
                 Barm=c(1,3), nsubject=252, pG_A1=0.8)

print(dtr13)