Introduction

ASSISTant is an R package for Adaptive Subgroup Selection In Sequential Trials. This vignette reproduces all the simulations in the original paper of Lai, Lavori and Liao (2014).

NOTE The number of simulations has been drastically reduced in this vignette in order to avoid taxing CRAN servers. The full_doc sources contain the complete sources and output; see files in the directory

system.file("full_doc", package="ASSISTant")
library(ASSISTant)
data(LLL.SETTINGS)
str(LLL.SETTINGS)
## List of 3
##  $ trialParameters:List of 4
##   ..$ N         : num [1:3] 300 400 500
##   ..$ type1Error: num 0.05
##   ..$ eps       : num 0.5
##   ..$ type2Error: num 0.2
##  $ scenarios      :List of 11
##   ..$ S0 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0 0 0 0 0 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S1 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.3 0 0.3 0 0.3 0 0.3 0 0.3 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S2 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.3 0 0.3 0 0.3 0 0.3 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S3 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.3 0 0.3 0 0 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S4 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0 0 0 0 0.3 0 0.3 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S5 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.6 0 0 0 0 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S6 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.5 0 0 0 0 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S7 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.5 0 0.4 0 0.3 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##   ..$ S8 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.5 0 0.4 0 0.3 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 2 1 1.5 1 1 1 0.5 1 0.5 ...
##   ..$ S9 :List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0.5 0 0.4 0 0.3 0 0 0 0 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 0.5 1 1 1 1.5 1 2 1 2 ...
##   ..$ S10:List of 2
##   .. ..$ mean: num [1:2, 1:6] 0 0 0 0 0 0 0 0.3 0 0.4 ...
##   .. ..$ sd  : num [1:2, 1:6] 1 1 1 1 1 1 1 1 1 1 ...
##  $ prevalences    :List of 2
##   ..$ table1: num [1:6] 0.167 0.167 0.167 0.167 0.167 ...
##   ..$ table2: num [1:6] 0.2 0.1 0.3 0.1 0.1 0.2

The LLL.SETTINGS list contains all the scenarios described in the paper.

Table 1 Results

Scenario S0

This is the null setting.

scenario <- LLL.SETTINGS$scenarios$S0
designParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
designA <- ASSISTDesign$new(trialParameters = LLL.SETTINGS$trialParameters,
                            designParameters = designParameters)
print(designA)
## Design Parameters:
##  Number of Groups: 6
##  Prevalence:
## 
##     Group1      Group2      Group3      Group4      Group5      Group6
## ----------  ----------  ----------  ----------  ----------  ----------
##  0.1666667   0.1666667   0.1666667   0.1666667   0.1666667   0.1666667
## 
##  Using Discrete Rankin scores? FALSE
## 
##  Normal Rankin Distribution means (null row, alt. row):
## 
## 
##         Group1   Group2   Group3   Group4   Group5   Group6
## -----  -------  -------  -------  -------  -------  -------
## Null         0        0        0        0        0        0
## Alt          0        0        0        0        0        0
## 
##  Normal Rankin Distribution SDs (null row, alt. row):
## 
## 
##         Group1   Group2   Group3   Group4   Group5   Group6
## -----  -------  -------  -------  -------  -------  -------
## Null         1        1        1        1        1        1
## Alt          1        1        1        1        1        1
## 
## Trial Parameters:
## List of 5
##  $ N         : num [1:3] 300 400 500
##  $ type1Error: num 0.05
##  $ eps       : num 0.5
##  $ type2Error: num 0.2
##  $ effectSize: num 0.0642
## 
## Boundaries:
## 
## 
##     btilde          b          c
## ----------  ---------  ---------
##  -1.460993   2.390404   2.491775
result <- designA$explore(numberOfSimulations = 50, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.020000; P(Reject H0_subgp) = 0.000000; P(Reject H0) = 0.020000
## P(Early stop for efficacy [futility]) = 0.020000 [0.640000]
## Mean [SD] Randomized N = 414.000000 [72.871147]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.20
##          2         0.46
##          3         0.34
## 
## Mean [SD] Lost N = 183.880000 [85.293644]
## Mean [SD] Analyzed N = 230.120000 [92.989716]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean          sd
## --------------  --------------  ----------  ----------
##              1               1   250.94118    4.955686
##              1               2   201.42857    6.241184
##              1               3   152.33333    3.511885
##              1               4   107.66667    6.027714
##              1               5    52.80000    4.266146
##              2               1   328.00000   12.727922
##              2               2   267.33333   25.026653
##              2               5    71.66667   15.143756
##              3               3   232.00000          NA
##              3               4   161.50000    9.192388
##              3               5    82.00000   12.529964
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      6       1         0.02
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1   17    7    3    3    5
##              2    2    3    0    0    3
##              3    0    0    1    2    3
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.98           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##         1              19               0
##         1              10               0
##         1               4               0
##         1               5               0
##         1              11               0
##         0               1               1
## NULL

Alternative Scenario S1

scenario <- LLL.SETTINGS$scenarios$S1
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.780000; P(Reject H0_subgp) = 0.040000; P(Reject H0) = 0.820000
## P(Early stop for efficacy [futility]) = 0.700000 [0.000000]
## Mean [SD] Randomized N = 372.000000 [90.441096]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.58
##          2         0.12
##          3         0.30
## 
## Mean [SD] Lost N = 45.500000 [112.449762]
## Mean [SD] Analyzed N = 326.500000 [94.545368]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean         sd
## --------------  --------------  ---------  ---------
##              1               5    48.0000         NA
##              2               1   330.0000         NA
##              3               1   415.6667   8.504900
##              3               3   246.0000         NA
##              3               5    80.8000   8.228001
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       1         0.02
##      3       1         0.02
##      6      39         0.78
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    0    0    0    0    1
##              2    1    0    0    0    0
##              3    3    0    1    0    5
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.18           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##      0.75               4               1
##      0.00               1               1
##      1.00               6               0
##      0.00              39              39
## NULL

Alternative Scenario S2

scenario <- LLL.SETTINGS$scenarios$S2
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.400000; P(Reject H0_subgp) = 0.320000; P(Reject H0) = 0.720000
## P(Early stop for efficacy [futility]) = 0.420000 [0.000000]
## Mean [SD] Randomized N = 436.000000 [82.709228]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.22
##          2         0.20
##          3         0.58
## 
## Mean [SD] Lost N = 122.160000 [123.479381]
## Mean [SD] Analyzed N = 313.840000 [100.538014]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   254.0000    2.828427
##              1               2   199.0000    8.485281
##              1               3   141.5000   21.920310
##              1               4    96.0000   11.313709
##              1               5    50.0000          NA
##              2               4   133.0000    5.656854
##              3               1   423.5000    7.778175
##              3               2   334.3333    7.505554
##              3               3   241.2500    6.701990
##              3               4   167.6667    5.267827
##              3               5    87.0000          NA
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       2         0.04
##      2       4         0.08
##      3       4         0.08
##      4       6         0.12
##      6      20         0.40
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    2    2    2    2    1
##              2    0    0    0    2    0
##              3    2    3    4    9    1
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.28           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.5000000               4               2
##  0.2000000               5               4
##  0.3333333               6               4
##  0.5384615              13               6
##  1.0000000               2               0
##  0.0000000              20              20
## NULL

Alternative Scenario S3

scenario <- LLL.SETTINGS$scenarios$S3
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.120000; P(Reject H0_subgp) = 0.420000; P(Reject H0) = 0.540000
## P(Early stop for efficacy [futility]) = 0.260000 [0.020000]
## Mean [SD] Randomized N = 460.000000 [69.985421]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.12
##          2         0.16
##          3         0.72
## 
## Mean [SD] Lost N = 223.540000 [126.875837]
## Mean [SD] Analyzed N = 236.460000 [122.725546]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean           sd
## --------------  --------------  ---------  -----------
##              1               1   251.6667    4.3204938
##              1               2   199.7500    7.8898669
##              1               3   142.0000    8.4852814
##              1               4   108.0000           NA
##              1               5    49.0000           NA
##              2               1   335.7500    6.7515430
##              2               2   267.0000           NA
##              2               3   209.0000           NA
##              2               5    65.0000           NA
##              3               1   416.8000    5.8480766
##              3               2   334.3750    9.8841504
##              3               3   261.6667   11.5902258
##              3               4   166.8000    9.8843310
##              3               5    82.5000    0.7071068
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1      12         0.24
##      2       6         0.12
##      3       2         0.04
##      5       1         0.02
##      6       6         0.12
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    6    4    2    1    1
##              2    4    1    1    0    1
##              3    5    8    3    5    2
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.46           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.2000000              15              12
##  0.5384615              13               6
##  0.6666667               6               2
##  1.0000000               6               0
##  0.7500000               4               1
##  0.0000000               6               6
## NULL

Alternative Scenario S4

scenario <- LLL.SETTINGS$scenarios$S4
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.060000; P(Reject H0_subgp) = 0.100000; P(Reject H0) = 0.160000
## P(Early stop for efficacy [futility]) = 0.040000 [0.180000]
## Mean [SD] Randomized N = 466.000000 [68.838765]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.12
##          2         0.10
##          3         0.78
## 
## Mean [SD] Lost N = 146.480000 [84.999457]
## Mean [SD] Analyzed N = 319.520000 [101.819015]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean          sd
## --------------  --------------  ----------  ----------
##              1               1   251.80000    1.483240
##              1               4   100.25000    8.811518
##              1               5    52.66667    1.527525
##              2               4   130.00000    8.366600
##              2               5    66.50000   24.748737
##              3               1   416.00000          NA
##              3               3   251.66667    8.524475
##              3               4   170.91667   11.712141
##              3               5    79.16667    5.980524
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       1         0.02
##      3       1         0.02
##      4       1         0.02
##      5       2         0.04
##      6       3         0.06
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    5    0    0    8    3
##              2    0    0    0    4    2
##              3    1    0    6   12    6
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.84           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.8333333               6               1
##  0.8333333               6               1
##  0.9583333              24               1
##  0.8181818              11               2
##  0.0000000               3               3
## NULL

Alternative Scenario S5

scenario <- LLL.SETTINGS$scenarios$S5
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.140000; P(Reject H0_subgp) = 0.660000; P(Reject H0) = 0.800000
## P(Early stop for efficacy [futility]) = 0.460000 [0.000000]
## Mean [SD] Randomized N = 444.000000 [67.491496]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.10
##          2         0.36
##          3         0.54
## 
## Mean [SD] Lost N = 270.120000 [141.724314]
## Mean [SD] Analyzed N = 173.880000 [126.469427]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   248.6923    5.807820
##              1               2   206.5000    2.121320
##              1               3   157.0000    4.242641
##              1               4   104.5000    4.949747
##              2               1   333.3333    4.457204
##              3               1   417.3125    5.313113
##              3               2   330.0000   12.727922
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1      30         0.60
##      2       1         0.02
##      3       1         0.02
##      4       1         0.02
##      6       7         0.14
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1   13    2    2    2    0
##              2    6    0    0    0    0
##              3   16    2    0    0    0
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##      0.2           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.1428571              35              30
##  0.7500000               4               1
##  0.5000000               2               1
##  0.5000000               2               1
##  0.0000000               7               7
## NULL

Alternative Scenario S6

scenario <- LLL.SETTINGS$scenarios$S6
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.060000; P(Reject H0_subgp) = 0.720000; P(Reject H0) = 0.780000
## P(Early stop for efficacy [futility]) = 0.480000 [0.040000]
## Mean [SD] Randomized N = 436.000000 [69.282032]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.12
##          2         0.40
##          3         0.48
## 
## Mean [SD] Lost N = 254.060000 [123.124196]
## Mean [SD] Analyzed N = 181.940000 [115.743808]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean          sd
## --------------  --------------  ----------  ----------
##              1               1   249.85714    7.336496
##              1               2   189.66667    5.507570
##              1               4   102.66667   11.718931
##              1               5    49.00000    8.485281
##              2               1   330.42857    6.106203
##              2               2   267.00000   14.142136
##              3               1   416.55556    3.574602
##              3               2   336.75000   12.685293
##              3               5    95.66667    7.023769
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1      26         0.52
##      2       6         0.12
##      4       1         0.02
##      5       3         0.06
##      6       3         0.06
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1   14    3    0    3    2
##              2    7    2    0    0    0
##              3    9    4    0    0    3
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.22           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.1333333              30              26
##  0.3333333               9               6
##  0.6666667               3               1
##  0.4000000               5               3
##  0.0000000               3               3
## NULL

Alternative Scenario S7

scenario <- LLL.SETTINGS$scenarios$S7
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.420000; P(Reject H0_subgp) = 0.480000; P(Reject H0) = 0.900000
## P(Early stop for efficacy [futility]) = 0.460000 [0.000000]
## Mean [SD] Randomized N = 428.000000 [85.809471]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.26
##          2         0.20
##          3         0.54
## 
## Mean [SD] Lost N = 137.180000 [132.051735]
## Mean [SD] Analyzed N = 290.820000 [102.494338]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean           sd
## --------------  --------------  ---------  -----------
##              1               1   243.5000   12.0208153
##              1               2   198.7500    5.5000000
##              1               3   153.5000    0.7071068
##              1               4    95.0000           NA
##              2               3   206.0000    8.4852814
##              3               1   422.0000           NA
##              3               2   341.0000    6.1644140
##              3               3   257.0000    5.0709255
##              3               4   165.6667    1.5275252
##              3               5    83.0000           NA
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       3         0.06
##      2       9         0.18
##      3       9         0.18
##      4       3         0.06
##      6      21         0.42
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    2    4    2    1    0
##              2    0    0    2    0    0
##              3    1    5    8    3    1
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##      0.1           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##      0.00               3               3
##      0.00               9               9
##      0.25              12               9
##      0.25               4               3
##      1.00               1               0
##      0.00              21              21
## NULL

Alternative Scenario S8

scenario <- LLL.SETTINGS$scenarios$S8
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.180000; P(Reject H0_subgp) = 0.500000; P(Reject H0) = 0.680000
## P(Early stop for efficacy [futility]) = 0.260000 [0.000000]
## Mean [SD] Randomized N = 460.000000 [72.843136]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.14
##          2         0.12
##          3         0.74
## 
## Mean [SD] Lost N = 170.720000 [104.977556]
## Mean [SD] Analyzed N = 289.280000 [85.082682]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   250.0000    5.291503
##              1               3   147.6000    2.509980
##              1               4   106.0000    6.055301
##              2               1   322.0000          NA
##              2               3   207.0000          NA
##              3               2   331.3333   11.590226
##              3               3   250.7647   10.195155
##              3               4   160.0000    1.732051
##              3               5    89.5000    3.109126
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       4         0.08
##      2       3         0.06
##      3      14         0.28
##      4       3         0.06
##      5       1         0.02
##      6       9         0.18
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    3    0    5    4    0
##              2    1    0    1    0    0
##              3    0    3   17    3    4
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.32           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.0000000               4               4
##  0.0000000               3               3
##  0.3913043              23              14
##  0.5714286               7               3
##  0.7500000               4               1
##  0.0000000               9               9
## NULL

Alternative Scenario S9

scenario <- LLL.SETTINGS$scenarios$S9
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.560000; P(Reject H0_subgp) = 0.360000; P(Reject H0) = 0.920000
## P(Early stop for efficacy [futility]) = 0.480000 [0.000000]
## Mean [SD] Randomized N = 426.000000 [85.260705]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.26
##          2         0.22
##          3         0.52
## 
## Mean [SD] Lost N = 125.840000 [153.421968]
## Mean [SD] Analyzed N = 300.160000 [127.468341]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   235.0000          NA
##              1               2   194.0000          NA
##              2               1   329.6667    2.081666
##              3               1   418.5000    4.949747
##              3               2   330.6250   13.394429
##              3               3   243.2500    7.365460
##              3               4   168.0000    1.414214
##              3               5    83.0000          NA
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       5         0.10
##      2       8         0.16
##      3       3         0.06
##      4       1         0.02
##      5       1         0.02
##      6      28         0.56
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    1    1    0    0    0
##              2    3    0    0    0    0
##              3    2    8    4    2    1
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.08           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.1666667               6               5
##  0.1111111               9               8
##  0.2500000               4               3
##  0.5000000               2               1
##  0.0000000               1               1
##  0.0000000              28              28
## NULL

Alternative Scenario S10

scenario <- LLL.SETTINGS$scenarios$S10
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table1,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50,
                          trueParameters = trueParameters, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.420000; P(Reject H0_subgp) = 0.000000; P(Reject H0) = 0.420000
## P(Early stop for efficacy [futility]) = 0.300000 [0.160000]
## Mean [SD] Randomized N = 422.000000 [91.003476]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.32
##          2         0.14
##          3         0.54
## 
## Mean [SD] Lost N = 78.100000 [118.835206]
## Mean [SD] Analyzed N = 343.900000 [117.819345]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean         sd
## --------------  --------------  ----------  ---------
##              1               1   252.00000         NA
##              1               5    50.42857   4.035556
##              3               1   418.25000   9.979145
##              3               2   331.00000         NA
##              3               5    81.00000   8.074652
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      6      21         0.42
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    1    0    0    0    7
##              2    0    0    0    0    0
##              3    4    1    0    0   16
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.58           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##         1               5               0
##         1               1               0
##         1              23               0
##         0              21              21
## NULL

Table 2 Results

Scenario S0

scenario <- LLL.SETTINGS$scenarios$S0
designParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
designA <- ASSISTDesign$new(trialParameters = LLL.SETTINGS$trialParameters,
                            designParameters = designParameters)
print(designA)
## Design Parameters:
##  Number of Groups: 6
##  Prevalence:
## 
##  Group1   Group2   Group3   Group4   Group5   Group6
## -------  -------  -------  -------  -------  -------
##     0.2      0.1      0.3      0.1      0.1      0.2
## 
##  Using Discrete Rankin scores? FALSE
## 
##  Normal Rankin Distribution means (null row, alt. row):
## 
## 
##         Group1   Group2   Group3   Group4   Group5   Group6
## -----  -------  -------  -------  -------  -------  -------
## Null         0        0        0        0        0        0
## Alt          0        0        0        0        0        0
## 
##  Normal Rankin Distribution SDs (null row, alt. row):
## 
## 
##         Group1   Group2   Group3   Group4   Group5   Group6
## -----  -------  -------  -------  -------  -------  -------
## Null         1        1        1        1        1        1
## Alt          1        1        1        1        1        1
## 
## Trial Parameters:
## List of 5
##  $ N         : num [1:3] 300 400 500
##  $ type1Error: num 0.05
##  $ eps       : num 0.5
##  $ type2Error: num 0.2
##  $ effectSize: num 0.0642
## 
## Boundaries:
## 
## 
##     btilde          b          c
## ----------  ---------  ---------
##  -1.460993   2.371573   2.467677
result <- designA$explore(numberOfSimulations = 50, showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.020000; P(Reject H0_subgp) = 0.020000; P(Reject H0) = 0.040000
## P(Early stop for efficacy [futility]) = 0.040000 [0.620000]
## Mean [SD] Randomized N = 406.000000 [79.308234]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.28
##          2         0.38
##          3         0.34
## 
## Mean [SD] Lost N = 175.400000 [80.752835]
## Mean [SD] Analyzed N = 230.600000 [105.285811]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   240.0000    9.518403
##              1               2   211.3333    7.364946
##              1               3   116.0000   16.970563
##              1               4    86.0000    7.071068
##              1               5    59.0000    3.162278
##              2               1   325.0000          NA
##              2               2   283.0000    4.242641
##              2               3   158.5000   14.849242
##              2               4   111.0000          NA
##              2               5    94.0000          NA
##              3               1   369.0000          NA
##              3               3   192.0000    7.071068
##              3               4   156.0000          NA
##              3               5   105.0000    7.937254
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      2       1         0.02
##      6       1         0.02
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1   11   12    2    5    5
##              2    1    2    2    1    1
##              3    1    0    2    1    3
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.96           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  1.0000000              13               0
##  0.9285714              14               1
##  1.0000000               6               0
##  1.0000000               7               0
##  1.0000000               9               0
##  0.0000000               1               1
## NULL

Alternative Scenario S1

scenario <- LLL.SETTINGS$scenarios$S1
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.780000; P(Reject H0_subgp) = 0.060000; P(Reject H0) = 0.840000
## P(Early stop for efficacy [futility]) = 0.760000 [0.000000]
## Mean [SD] Randomized N = 366.000000 [84.780629]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.58
##          2         0.18
##          3         0.24
## 
## Mean [SD] Lost N = 44.200000 [97.398403]
## Mean [SD] Analyzed N = 321.800000 [67.726340]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean         sd
## --------------  --------------  ---------  ---------
##              2               2   282.0000         NA
##              3               2   347.6667   4.725816
##              3               3   205.0000         NA
##              3               4   133.5000   3.535534
##              3               5   103.2500   1.707825
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      2       1         0.02
##      3       1         0.02
##      5       1         0.02
##      6      39         0.78
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    0    0    0    0    0
##              2    0    1    0    0    0
##              3    0    3    1    2    4
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.16           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##      0.75               4               1
##      0.00               1               1
##      1.00               2               0
##      0.75               4               1
##      0.00              39              39
## NULL

Alternative Scenario S2

scenario <- LLL.SETTINGS$scenarios$S2
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.520000; P(Reject H0_subgp) = 0.220000; P(Reject H0) = 0.740000
## P(Early stop for efficacy [futility]) = 0.520000 [0.020000]
## Mean [SD] Randomized N = 424.000000 [79.693290]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.22
##          2         0.32
##          3         0.46
## 
## Mean [SD] Lost N = 79.800000 [98.886866]
## Mean [SD] Analyzed N = 344.200000 [80.416264]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean          sd
## --------------  --------------  ----------  ----------
##              1               1   237.00000          NA
##              1               2   222.00000          NA
##              1               3   122.00000          NA
##              1               4    87.00000    5.567764
##              2               2   281.00000          NA
##              2               4   137.00000          NA
##              3               2   353.00000    4.242641
##              3               3   198.00000   26.870058
##              3               4   149.00000   11.302655
##              3               5    95.66667    5.686241
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       1         0.02
##      2       1         0.02
##      4       8         0.16
##      5       1         0.02
##      6      26         0.52
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    1    1    1    3    0
##              2    0    1    0    1    0
##              3    0    2    2    9    3
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.26           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.0000000               1               1
##  0.7500000               4               1
##  1.0000000               3               0
##  0.3846154              13               8
##  0.6666667               3               1
##  0.0000000              26              26
## NULL

Alternative Scenario S3

scenario <- LLL.SETTINGS$scenarios$S3
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.060000; P(Reject H0_subgp) = 0.560000; P(Reject H0) = 0.620000
## P(Early stop for efficacy [futility]) = 0.300000 [0.040000]
## Mean [SD] Randomized N = 458.000000 [64.174507]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.08
##          2         0.26
##          3         0.66
## 
## Mean [SD] Lost N = 244.860000 [104.019035]
## Mean [SD] Analyzed N = 213.140000 [90.550022]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean           sd
## --------------  --------------  ---------  -----------
##              1               1   240.1250    7.7170960
##              1               2   209.8000    7.5099933
##              1               4   100.0000           NA
##              2               1   323.0000    4.2426407
##              2               2   277.5000    6.3639610
##              2               4   116.0000           NA
##              2               5    75.0000           NA
##              3               1   401.5000    2.1213203
##              3               2   345.6667   12.6923299
##              3               3   187.5000    0.7071068
##              3               4   151.0000           NA
##              3               5   109.0000    1.4142136
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       9         0.18
##      2      18         0.36
##      4       1         0.02
##      6       3         0.06
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    8   10    0    1    0
##              2    2    2    0    1    1
##              3    2   15    2    1    2
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.38           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.2500000              12               9
##  0.3333333              27              18
##  1.0000000               2               0
##  0.6666667               3               1
##  1.0000000               3               0
##  0.0000000               3               3
## NULL

Alternative Scenario S4

scenario <- LLL.SETTINGS$scenarios$S4
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.160000; P(Reject H0_subgp) = 0.120000; P(Reject H0) = 0.280000
## P(Early stop for efficacy [futility]) = 0.140000 [0.160000]
## Mean [SD] Randomized N = 452.000000 [78.869564]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.18
##          2         0.12
##          3         0.70
## 
## Mean [SD] Lost N = 103.560000 [60.185143]
## Mean [SD] Analyzed N = 348.440000 [65.372140]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup        mean          sd
## --------------  --------------  ----------  ----------
##              1               1   238.00000          NA
##              1               3   127.00000    8.485281
##              1               4    96.33333    8.594572
##              1               5    61.50000    9.071935
##              3               3   196.00000   22.538855
##              3               4   151.64286    7.601692
##              3               5   102.80000   10.347302
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      4       5         0.10
##      5       1         0.02
##      6       8         0.16
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    1    0    2    6    6
##              2    0    0    0    0    0
##              3    0    0    3   14   10
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.72           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##    1.0000               1               0
##    1.0000               5               0
##    0.7500              20               5
##    0.9375              16               1
##    0.0000               8               8
## NULL

Alternative Scenario S5

scenario <- LLL.SETTINGS$scenarios$S5
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.100000; P(Reject H0_subgp) = 0.740000; P(Reject H0) = 0.840000
## P(Early stop for efficacy [futility]) = 0.400000 [0.000000]
## Mean [SD] Randomized N = 446.000000 [73.429127]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.14
##          2         0.26
##          3         0.60
## 
## Mean [SD] Lost N = 282.580000 [122.120230]
## Mean [SD] Analyzed N = 163.420000 [104.629783]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   238.8182    7.769404
##              1               2   200.3333   10.503967
##              2               1   330.0000    9.183318
##              2               2   278.0000          NA
##              2               4   113.0000          NA
##              3               1   399.2667   10.746871
##              3               2   357.6000    6.655824
##              3               3   211.5000    3.535534
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1      28         0.56
##      2       8         0.16
##      3       1         0.02
##      6       5         0.10
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1   11    3    0    0    0
##              2    7    1    0    1    0
##              3   15    5    2    0    0
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.16           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.1515152              33              28
##  0.1111111               9               8
##  0.5000000               2               1
##  1.0000000               1               0
##  0.0000000               5               5
## NULL

Alternative Scenario S6

scenario <- LLL.SETTINGS$scenarios$S6
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.080000; P(Reject H0_subgp) = 0.680000; P(Reject H0) = 0.760000
## P(Early stop for efficacy [futility]) = 0.320000 [0.000000]
## Mean [SD] Randomized N = 456.000000 [70.450446]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.12
##          2         0.20
##          3         0.68
## 
## Mean [SD] Lost N = 275.360000 [129.742722]
## Mean [SD] Analyzed N = 180.640000 [114.948215]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   236.8750    7.453427
##              1               2   199.5000   13.435029
##              1               3   126.0000          NA
##              1               4   106.0000          NA
##              1               5    61.0000          NA
##              2               1   326.8000    6.648308
##              2               4   134.0000          NA
##              3               1   395.0526    9.507153
##              3               2   359.0000   14.000000
##              3               3   204.3333    8.504900
##              3               4   132.0000          NA
##              3               5    85.0000          NA
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1      26         0.52
##      2       5         0.10
##      3       3         0.06
##      6       4         0.08
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    8    2    1    1    1
##              2    5    0    0    1    0
##              3   19    3    3    1    1
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.24           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##    0.1875              32              26
##    0.0000               5               5
##    0.2500               4               3
##    1.0000               3               0
##    1.0000               2               0
##    0.0000               4               4
## NULL

Alternative Scenario S7

scenario <- LLL.SETTINGS$scenarios$S7
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.580000; P(Reject H0_subgp) = 0.380000; P(Reject H0) = 0.960000
## P(Early stop for efficacy [futility]) = 0.540000 [0.000000]
## Mean [SD] Randomized N = 416.000000 [86.567185]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.30
##          2         0.24
##          3         0.46
## 
## Mean [SD] Lost N = 104.820000 [132.987446]
## Mean [SD] Analyzed N = 311.180000 [105.744144]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean         sd
## --------------  --------------  ---------  ---------
##              1               2   212.5000    5.00000
##              1               3   123.0000         NA
##              2               2   290.0000         NA
##              3               1   393.0000         NA
##              3               2   345.2000   10.94075
##              3               3   206.5556   12.74864
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       1         0.02
##      2      10         0.20
##      3       8         0.16
##      6      29         0.58
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    0    4    1    0    0
##              2    0    1    0    0    0
##              3    1    5    9    0    0
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.04           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##  coverage   selectedCount   rejectedCount
## ---------  --------------  --------------
##       0.0               1               1
##       0.0              10              10
##       0.2              10               8
##       0.0              29              29
## NULL

Alternative Scenario S8

scenario <- LLL.SETTINGS$scenarios$S8
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.360000; P(Reject H0_subgp) = 0.480000; P(Reject H0) = 0.840000
## P(Early stop for efficacy [futility]) = 0.400000 [0.000000]
## Mean [SD] Randomized N = 438.000000 [83.029378]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.22
##          2         0.18
##          3         0.60
## 
## Mean [SD] Lost N = 121.880000 [114.118563]
## Mean [SD] Analyzed N = 316.120000 [89.724440]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup    mean          sd
## --------------  --------------  ------  ----------
##              1               2   209.0   12.727922
##              1               3   120.0    5.385165
##              1               4    87.0          NA
##              3               1   393.0          NA
##              3               2   350.5   11.030261
##              3               3   198.0   11.162437
##              3               4   153.5    4.203173
##              3               5   100.5    9.110434
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       1         0.02
##      2       5         0.10
##      3      11         0.22
##      4       3         0.06
##      5       4         0.08
##      6      18         0.36
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    0    2    5    1    0
##              2    0    0    0    0    0
##              3    1    4   11    4    4
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.16           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.0000000               1               1
##  0.1666667               6               5
##  0.3125000              16              11
##  0.4000000               5               3
##  0.0000000               4               4
##  0.0000000              18              18
## NULL

Alternative Scenario S9

scenario <- LLL.SETTINGS$scenarios$S9
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.540000; P(Reject H0_subgp) = 0.420000; P(Reject H0) = 0.960000
## P(Early stop for efficacy [futility]) = 0.480000 [0.000000]
## Mean [SD] Randomized N = 426.000000 [85.260705]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.26
##          2         0.22
##          3         0.52
## 
## Mean [SD] Lost N = 124.200000 [148.980207]
## Mean [SD] Analyzed N = 301.800000 [115.994370]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   234.0000          NA
##              1               2   219.5000    2.121320
##              1               5    65.0000          NA
##              2               2   288.0000          NA
##              3               1   384.0000    2.828427
##              3               2   346.6667   15.491933
##              3               3   195.4000    7.300685
##              3               4   159.5000   10.606602
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      1       3         0.06
##      2      11         0.22
##      3       5         0.10
##      4       2         0.04
##      6      27         0.54
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    1    2    0    0    1
##              2    0    1    0    0    0
##              3    2    9    5    2    0
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##     0.04           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  0.0000000               3               3
##  0.0833333              12              11
##  0.0000000               5               5
##  0.0000000               2               2
##  1.0000000               1               0
##  0.0000000              27              27
## NULL

Alternative Scenario S10

scenario <- LLL.SETTINGS$scenarios$S10
trueParameters <- list(prevalence = LLL.SETTINGS$prevalences$table2,
                       mean = scenario$mean,
                       sd = scenario$sd)
result <- designA$explore(numberOfSimulations = 50, trueParameters = trueParameters,
                          showProgress = FALSE)
analysis <- designA$analyze(result)
print(designA$summary(analysis))
## P(Reject H0_ITT) = 0.280000; P(Reject H0_subgp) = 0.020000; P(Reject H0) = 0.300000
## P(Early stop for efficacy [futility]) = 0.260000 [0.280000]
## Mean [SD] Randomized N = 420.000000 [83.299313]
## 
## Stage at exit (proportion)
## 
## 
##  exitStage   proportion
## ----------  -----------
##          1         0.26
##          2         0.28
##          3         0.46
## 
## Mean [SD] Lost N = 105.280000 [101.708876]
## Mean [SD] Analyzed N = 314.720000 [98.752430]
## 
## Mean loss by futility stage and subgroup
## 
## 
##  FutilityStage   selectedGroup       mean          sd
## --------------  --------------  ---------  ----------
##              1               1   237.0000    5.244044
##              1               2   215.0000    8.869423
##              1               5    56.5000    7.148426
##              2               1   325.0000          NA
##              2               5    84.0000          NA
##              3               1   400.0000          NA
##              3               2   359.0000          NA
##              3               5   100.7059   10.042190
## 
## Chance of each subpopulation rejected
## 
## 
##  group   count   proportion
## ------  ------  -----------
##      5       1         0.02
##      6      14         0.28
## 
## Counts by futility stage and subgroup choice
## 
## 
##  FutilityStage   G1   G2   G3   G4   G5
## --------------  ---  ---  ---  ---  ---
##              1    5    4    0    0    6
##              2    1    0    0    0    1
##              3    1    1    0    0   17
## 
## CI Statistics:
## Overall coverage and coverage for rejections:
## 
##  overall   rejection
## --------  ----------
##      0.7           0
## 
## P(theta_test is in the confidence interval)
## 
## 
##   coverage   selectedCount   rejectedCount
## ----------  --------------  --------------
##  1.0000000               7               0
##  1.0000000               5               0
##  0.9583333              24               1
##  0.0000000              14              14
## NULL

References

Lai, Tze Leung, Philip W. Lavori, and Olivia Yueh-Wen Liao. 2014. “Adaptive Choice of Patient Subgroup for Comparing Two Treatments.” Contemporary Clinical Trials 39 (2): 191–200. doi:10.1016/j.cct.2014.09.001.