Type: Package
Title: Sequential Multiple Assignment Randomized Trial Design
Version: 0.74
Date: 2024-03-04
Description: SMART trial design, as described by He, J., McClish, D., Sabo, R. (2021) <doi:10.1080/19466315.2021.1883472>, includes multiple stages of randomization, where participants are randomized to an initial treatment in the first stage and then subsequently re-randomized between treatments in the following stage.
License: GPL (≥ 3)
Depends: R (≥ 4.1.0), methods, graphics, stats
Imports: knitr
NeedsCompilation: no
RoxygenNote: 7.1.0
VignetteBuilder: knitr
URL: https://cran.r-project.org/package=smartDesign
Suggests: testthat, rmarkdown
Packaged: 2024-03-04 18:04:38 UTC; sinnwell
Author: Jason Sinnwell ORCID iD [aut, cre], Jun(Jessie) He ORCID iD [aut], Abraham Eyman Casey [aut]
Maintainer: Jason Sinnwell <sinnwell.jason@mayo.edu>
Repository: CRAN
Date/Publication: 2024-03-04 19:30:02 UTC

Power Dynamic Treatment Regimen (DTR) Trial design clinical trial calculations

Description

Power Calculations Comparing two Dynamic Treatment Regimen (DTR) Trial design clinical trial calculations

Usage

powerDTR(dtr1, dtr2, pG_A1 = 0.8, pG_A2 = 0.8, alpha=0.05)

Arguments

dtr1

an object of smartDTR class, created by function of the same name

dtr2

an object of smartDTR class, created by function of the same name

pG_A1

probability of response to therapy given assignment to A1

pG_A2

probability of response to therapy given assignment to A2

alpha

accepted type-I error rate for power calculations

Details

more details on power DTR

Value

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

powerdat:

data.frame with sens, spec, mu, sigsq and sample size, power

Author(s)

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

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)

  mumat24 <- cbind(G1=c(25,32), G0=c(18,23))
  varmat24 <- cbind(G1=c(100,100),G0=c(100,100))

  dtr24 <- smartDTR(mu_Barm=mumat24, sigsq_Barm=varmat24,
                   Barm=c(2,4), nsubject=252, pG_A1=0.8, pG_A2=0.8)

  pdtr13vs24 <- powerDTR(dtr13, dtr24)
  print(pdtr13vs24)  ## plot(pdtr13vs24)

Power for Single Sequential Treatment (SST) Trial design clinical trial calculations

Description

Power Calculations Comparing two Single Sequential Treatment Treatment (SST) Trial design clinical trial calculations

Usage

powerSST(sst1, sst2, pG_A1 = 0.8, pG_A2 = 0.8, alpha=0.05)

Arguments

sst1

an object of smartSST class, created by function of the same name

sst2

an object of smartSST class, created by function of the same name

pG_A1

probability of response to therapy given assignment to A1

pG_A2

probability of response to therapy given assignment to A2

alpha

accepted type-I error rate for power calculations

Details

more details to come

Value

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

powerdat:

data.frame with sens, spec, mu, sigsq and sample size, power

Author(s)

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

Examples

  sst1 <- smartSST(mu_Barm=c(G1=30, G0=20), sigsq_Barm=c(G1=16,G0=16),
         Barm=1, sens=seq(.6, 1, by=.1),  spec=seq(.6, 1, by=.1),
         nsubject=252)
  sst2 <- smartSST(mu_Barm=c(G1=20, G0=30), sigsq_Barm=c(G1=16,G0=16),
         Barm=2, sens=seq(.6, 1, by=.1),  spec=seq(.6, 1, by=.1),
         nsubject=252)

  psst12 <- powerSST(sst1, sst2)
  print(psst12) ## plot(psst12)

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)

Single Sequential Trial design clinical trial calculations

Description

Single Sequential Trial design clinical trial calculations

Usage

smartSST(mu_Barm=c(G1=30, G0=20), sigsq_Barm=c(G1=100, G0=100),
                     nsubject=500,
                     Barm=1, 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 = 0.5)

Arguments

mu_Barm

named vector of the means for the Barm for the smart SST trial, with names 'G1' and 'G0'

sigsq_Barm

named vector of the variances (sigma-squared) for the Barm for the smart SST trial, with names 'G1' and 'G0'

nsubject

total sample size for the trial

Barm

for the second phase of the trial, the 'B' level for which the 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 Barm

Details

more details on smart SST

Value

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

sstdat:

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

mu_Barm:

The value of mu_Barm passed to the function

sigsq_Barm:

The value of sigsq_Barm passed to the function

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

sst1 <- smartSST(mu_Barm=c(G1=30, G0=20), sigsq_Barm=c(G1=16,G0=16),
        Barm=1, sens=seq(.6, 1, by=.1),  spec=seq(.6, 1, by=.1),
        nsubject=252)
print(sst1$sstdat, digits=2)