issue when using restriction with 10 outcomes about buysetest

[bozenne]

Hello Samuel,

Humm, I tried to replicate the issue without success:

 set.seed(1)
    dt <- simBuyseTest(n.T = 50, n.C = 50,
                       names.strata = "strat_column", n.strata = 3,
                       argsTTE = list(name = c("tte1","tte2","tte3"), name.censoring = c("cnsr1","cnsr2","cnsr3"),
                                      scale.T = c(200,100,250), scale.censoring.T = c(10^5,10^5,10^5),
                                      scale.C = c(200,100,250), scale.censoring.C = c(10^5,10^5,10^5)),
                       argsBin = list(name = "bin_var"))

    formula1 <- treatment ~ strat_column + tte(tte1, status = cnsr1, threshold = 10, restriction = 365) + tte(tte2, status = cnsr2, threshold = 10, restriction = 365) + bin(bin_var, operator = "<0") + tte(tte3, status = cnsr3, threshold = 10, restriction = 365)

    GPC.v1 <- BuyseTest(formula1,
                        data = dt,
                        trace = FALSE)

    GPC.v2 <- BuyseTest(treatment = "treatment",
                        strata = "strat_column",
                        endpoint = c("tte1","tte2","bin_var","tte3"),
                        status = c("cnsr1","cnsr2","cnsr3"),
                        type = c("tte","tte","bin","tte"),
                        operator = c(">0",">0","<0",">0"),
                        threshold = c(10,10,NA,10),
                        restriction = c(365,365,NA,365),
                        data = dt,
                        trace = FALSE)

Could you provide a reproducible example so I can debug it?
Thanks!
Brice

[SamSalvaggio]

Thanks for your response.

Here is the code, it seems it works when specifying the formula similarly to "GPC.v2", however it returns the same error message when using the formula1 variable in GPC.v1

Code :

set.seed(1)
dt <- simBuyseTest(n.T = 50, n.C = 50,
                   names.strata = "strat_column", n.strata = 6,
                   argsTTE = list(name = c("tte1","tte2","tte3", "tte4", "tte5", "tte6", "tte7", "tte8", "tte9"), 
                                  name.censoring = c("cnsr1","cnsr2","cnsr3", "cnsr4", "cnsr5", "cnsr6", "cnsr7", "cnsr8", "cnsr9"),
                                  scale.T = c(200,100,250,300,150,200,350,400,450), scale.censoring.T = c(10^5,10^5,10^5,10^5,10^5,10^5,10^5,10^5,10^5),
                                  scale.C = c(200,100,250,300,150,200,350,400,450), scale.censoring.C = c(10^5,10^5,10^5,10^5,10^5,10^5,10^5,10^5,10^5)),
                   argsBin = list(name = "bin_var"))

formula1 <- treatment ~ strat_column + 
  tte(tte1, status = cnsr1, threshold = 10, restriction = 365) + 
  tte(tte2, status = cnsr2, threshold = 10, restriction = 365) + 
  tte(tte3, status = cnsr3, threshold = 10, restriction = 365) +
  tte(tte4, status = cnsr4, threshold = 10, restriction = 365) +
  bin(bin_var, operator = "<0") + 
  tte(tte5, status = cnsr5, threshold = 10, restriction = 365) + 
  tte(tte6, status = cnsr6, threshold = 10, restriction = 365) + 
  tte(tte7, status = cnsr7, threshold = 10, restriction = 365) +
  tte(tte8, status = cnsr8, threshold = 10, restriction = 365) +
  tte(tte9, status = cnsr9, threshold = 10, restriction = 365)

GPC.v1 <- BuyseTest(formula1,
                    data = dt,
                    trace = FALSE)

GPC.v2 <- BuyseTest(treatment = "treatment",
                    strata = "strat_column",
                    endpoint = c("tte1","tte2","tte3","tte4","bin_var","tte5","tte6","tte7","tte8","tte9"),
                    status = c("cnsr1","cnsr2","cnsr3","cnsr4","cnsr5","cnsr6","cnsr7","cnsr8","cnsr9"),
                    type = c("tte","tte","tte","tte","bin","tte","tte","tte","tte","tte"),
                    operator = c(">0",">0",">0",">0","<0",">0",">0",">0",">0",">0"),
                    threshold = c(10,10,10,10,NA,10,10,10,10,10),
                    restriction = c(365,365,365,365,NA,365,365,365,365,365),
                    data = dt,
                    trace = FALSE)

[bozenne]

Thanks! I can reproduce the error now. I'll have a look and will fix it - probably only next week though

[bozenne]

Fixed in the new version 2.4.3