Skip to contents

The DST01 Disposition Table provides an overview of patients study completion. For patients who discontinued the study a reason is provided.

Usage

dst01_main(
  adam_db,
  arm_var = "ARM",
  lbl_overall = "All {Patient_label}",
  study_status_var = "EOSSTT",
  detail_vars = list(Discontinued = c("DCSREAS")),
  trt_status_var = NULL,
  ...
)

dst01_pre(adam_db, ...)

dst01_post(tlg, prune_0 = TRUE, ...)

dst01

Format

An object of class chevron_t of length 1.

Arguments

adam_db

(list of data.frames) object containing the ADaM datasets

arm_var

(string) variable. Usually one of ARM, ACTARM, TRT01A, or TRT01A.

lbl_overall

(string) label used for overall column, if set to NULL the overall column is omitted

study_status_var

(string) variable used to define patient status. Default is EOSSTT, however can also be a variable name with the pattern EOPxxSTT where xx must be substituted by 2 digits referring to the analysis period.

detail_vars

Named (list) of grouped display of study_status_var. The names must be subset of unique levels of study_status_var.

trt_status_var

(string) variable of treatment status.

...

not used.

tlg

(TableTree, Listing or ggplot) object typically produced by a main function.

prune_0

(flag) remove 0 count rows

Value

the main function returns an rtables object.

the preprocessing function returns a list of data.frame.

the postprocessing function returns an rtables object or an ElementaryTable (null report).

Details

  • Default patient disposition table summarizing the reasons for patients withdrawal.

  • Numbers represent absolute numbers of patients and fraction of N.

  • Remove zero-count rows.

  • Split columns by arm.

  • Include a total column by default.

  • Sort withdrawal reasons by alphabetic order.

Functions

  • dst01_main(): Main TLG function

  • dst01_pre(): Preprocessing

  • dst01_post(): Postprocessing

Note

  • adam_db object must contain an adsl table with the columns specified by status_var and disc_reason_var.

Examples

run(dst01, syn_data, detail_vars = list(Ongoing = "STDONS"))
#>               A: Drug X    B: Placebo   C: Combination   All Patients
#>                 (N=15)       (N=15)         (N=15)          (N=45)   
#>   ———————————————————————————————————————————————————————————————————
#>   Completed   10 (66.7%)   10 (66.7%)     10 (66.7%)      30 (66.7%) 

run(dst01, syn_data, detail_vars = list(Discontinued = "DCSREAS", Ongoing = "STDONS"))
#>                                     A: Drug X    B: Placebo   C: Combination   All Patients
#>                                       (N=15)       (N=15)         (N=15)          (N=45)   
#>   —————————————————————————————————————————————————————————————————————————————————————————
#>   Completed                         10 (66.7%)   10 (66.7%)     10 (66.7%)      30 (66.7%) 
#>   Discontinued                      5 (33.3%)    5 (33.3%)      5 (33.3%)       15 (33.3%) 
#>     ADVERSE EVENT                       0            0           1 (6.7%)        1 (2.2%)  
#>     DEATH                           2 (13.3%)    4 (26.7%)      3 (20.0%)       9 (20.0%)  
#>     LACK OF EFFICACY                2 (13.3%)        0              0            2 (4.4%)  
#>     PHYSICIAN DECISION                  0            0           1 (6.7%)        1 (2.2%)  
#>     PROTOCOL VIOLATION                  0         1 (6.7%)          0            1 (2.2%)  
#>     WITHDRAWAL BY PARENT/GUARDIAN    1 (6.7%)        0              0            1 (2.2%)  

run(
  dst01, syn_data,
  detail_vars = list(
    Discontinued = c("DCSREASGP", "DCSREAS"),
    Ongoing = "STDONS"
  )
)
#>                                       A: Drug X    B: Placebo   C: Combination   All Patients
#>                                         (N=15)       (N=15)         (N=15)          (N=45)   
#>   ———————————————————————————————————————————————————————————————————————————————————————————
#>   Completed                           10 (66.7%)   10 (66.7%)     10 (66.7%)      30 (66.7%) 
#>   Discontinued                        5 (33.3%)    5 (33.3%)      5 (33.3%)       15 (33.3%) 
#>     Safety                                                                                   
#>       ADVERSE EVENT                       0            0           1 (6.7%)        1 (2.2%)  
#>       DEATH                           2 (13.3%)    4 (26.7%)      3 (20.0%)       9 (20.0%)  
#>     Non-Safety                                                                               
#>       LACK OF EFFICACY                2 (13.3%)        0              0            2 (4.4%)  
#>       PHYSICIAN DECISION                  0            0           1 (6.7%)        1 (2.2%)  
#>       PROTOCOL VIOLATION                  0         1 (6.7%)          0            1 (2.2%)  
#>       WITHDRAWAL BY PARENT/GUARDIAN    1 (6.7%)        0              0            1 (2.2%)