FSTG01 Subgroup Analysis of Best Overall Response.

Usage

fstg01_main(
  adam_db,
  dataset = "adrs",
  arm_var = "ARM",
  rsp_var = "is_rsp",
  subgroups = c("SEX", "AGEGR1", "RACE"),
  strata_var = NULL,
  ...
)

fstg01_pre(adam_db, ...)

fstg01_post(tlg, ...)

fstg01

Format

An object of class chevron_g of length 1.

Arguments

adam_db

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

dataset

(string) the name of a table in the adam_db object.

arm_var

(string) the arm variable name used for group splitting.

rsp_var

(string) the response variable name to flag whether each subject is a binary response or not.

subgroups

(character) the subgroups variable name to list baseline risk factors.

strata_var

(character) required if stratified analysis is performed.

...

Further arguments passed to g_forest and extract_rsp_subgroups (a wrapper for h_odds_ratio_subgroups_df and h_proportion_subgroups_df). For details, see the documentation in tern. Commonly used arguments include col_symbol_size, col, vline, groups_lists, conf_level, method, label_all, etc.

tlg

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

Value

a gTree object.

Details

• No overall value.

• Keep zero count rows by default.

Functions

• fstg01_main(): Main TLG Function

• fstg01_pre(): Preprocessing

• fstg01_post(): Postprocessing

Note

• adam_db object must contain the table specified by dataset with "PARAMCD", "ARM", "AVALC", and the columns specified by subgroups which is denoted as c("SEX", "AGEGR1", "RACE") by default.

Examples

library(dplyr)
library(dunlin)

proc_data <- log_filter(
  syn_data,
  PARAMCD == "BESRSPI" & ARM %in% c("A: Drug X", "B: Placebo"), "adrs"
)
run(fstg01, proc_data,
  subgroups = c("SEX", "AGEGR1", "RACE"),
  conf_level = 0.90, dataset = "adrs"
)

#> gTree[GRID.gTree.11]