Adding new module using summarize_glm_count

[llrs-roche]

Pull Request

Fixes #1379

This PR adds a new module tm_t_count (name open to improvements) that uses tern::summarize_glm_count to create a table output per request.

Sample app on how to use it

devtools::load_all("../teal.modules.clinical")
data <- within(teal_data(), {
  library("tern")
  ADSL <- tern_ex_adsl
  ADTTE <- tern_ex_adtte
})

join_keys(data) <- default_cdisc_join_keys[names(data)]

arm_ref_comp <- list(
  ACTARMCD = list(
    ref = "ARM B",
    comp = c("ARM A", "ARM C")
  ),
  ARM = list(
    ref = "B: Placebo",
    comp = c("A: Drug X", "C: Combination")
  )
)

ADSL <- data[["ADSL"]]
ADTTE <- data[["ADTTE"]]
# Initialize the teal app
app <- init(
  data = data,
  modules = modules(
    tm_t_glm_counts(
      dataname = "ADTTE",
      arm_var = choices_selected(
        variable_choices(ADTTE, c("ARM", "ARMCD", "ACTARMCD")),
        "ARMCD"
      ),
      arm_ref_comp = arm_ref_comp,
      aval_var = choices_selected(
        variable_choices(ADTTE, "AVAL"),
        "AVAL"
      ),
      strata_var = choices_selected(
        variable_choices(ADSL, "SEX"),
        NULL
      ),
      offset_var = choices_selected(
        variable_choices(ADSL, "AGE"),
        NULL
      ),
      cov_var = choices_selected(
        variable_choices(ADTTE, "SITEID"),
        NULL
      )
    )
  )
)

if (interactive()) {
  shinyApp(ui = app$ui, server = app$server)
}

Image of the output

[llrs-roche]

I would appreciate feedback on:

• Documentation of the module: Besides the tern function used, is it clear?
• Code shown from the app don't use the magrittr pipe and variables are overwritten.
• Input validation and the optional arguments reset to NULL.
• No tests yet: What would be of interest to test?

[github-actions]

Unit Tests Summary

  1 files   71 suites   13s ⏱️
729 tests 153 ✅ 576 💤 0 ❌
869 runs  174 ✅ 695 💤 0 ❌

Results for commit e178988.

♻️ This comment has been updated with latest results.

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
tm_t_glm_counts	👶		$+0.34$	$+2$	$+2$	$0$	$0$

Additional test case details

Test Suite	$Status$	Time on main	$±Time$	Test Case
examples	👶		$+0.02$	example_tm_t_glm_counts.Rd
tm_t_glm_counts	👶		$+0.12$	e2e_tm_t_events_by_grade_Selecting_arm_var_changes_the_table_and_does_not_throw_validation_errors.
tm_t_glm_counts	👶		$+0.22$	e2e_tm_t_glm_counts_Module_initializes_in_teal_without_errors_and_produces_table_output.

Results for commit 6130ced

♻️ This comment has been updated with latest results.

[m7pr]

Hey, for me it looks fine.
I left comments on how to update the documentation to make it clearer what is happening potentially.

You should clarify with user about this startified/unstratified output, and if it's not needed, just delete from the code.

Also please remove commented parameters if you don't use them.

[m7pr]

for the tests, just check out other modules and their shinytest2 tests - just test if the specific inputs are visible and the output is visible. follow some tests for another module that has tables

[llrs-roche]

Thanks for the feedback. I have added a couple of tests too