Expand get_stats to cover all analysis functions (#1123)

Fixes #1108

NAMESPACE

@@ -288,6 +288,7 @@ export(tabulate_survival_biomarkers)
 export(tabulate_survival_subgroups)
 export(tern_default_formats)
 export(tern_default_labels)
+export(tern_default_stats)
 export(test_proportion_diff)
 export(to_n)
 export(to_string_matrix)

NEWS.md

@@ -7,6 +7,7 @@
 * Added internal utility function `apply_auto_formatting` to check for `"auto"` formats and replace them with 
   implementation of `format_auto` in analyze functions.
 * Added utility function `labels_use_control` to modify labels with control specifications.
+* Added list containing default statistics for each method group, `tern_default_stats`.
 
 ### Enhancements
 * Added `ref_group_coxph` parameter to `g_km` to specify the reference group used for pairwise Cox-PH calculations when `annot_coxph = TRUE`.

R/abnormal.R

@@ -13,6 +13,8 @@
 #' @param abnormal (named `list`)\cr list identifying the abnormal range level(s) in `var`. Defaults to
 #'   `list(Low = "LOW", High = "HIGH")` but you can also group different levels into the named list,
 #'   for example, `abnormal = list(Low = c("LOW", "LOW LOW"), High = c("HIGH", "HIGH HIGH"))`.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("abnormal")`
+#'   to see available statistics for this function.
 #'
 #' @note
 #' * `count_abnormal()` only works with a single variable containing multiple abnormal levels.

R/abnormal_by_baseline.R

@@ -18,6 +18,8 @@
 #'
 #' @inheritParams argument_convention
 #' @param abnormal (`character`)\cr identifying the abnormal range level(s) in `.var`.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("abnormal_by_baseline")`
+#'   to see available statistics for this function.
 #'
 #' @note
 #' * `df` should be filtered to include only post-baseline records.

R/abnormal_by_marked.R

@@ -14,6 +14,8 @@
 #' @inheritParams argument_convention
 #' @param category (`list`)\cr with different marked category names for single
 #'   and last or replicated.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("abnormal_by_marked")`
+#'   to see available statistics for this function.
 #'
 #' @note `Single, not last` and `Last or replicated` levels are mutually exclusive. If a patient has
 #'   abnormalities that meet both the `Single, not last` and `Last or replicated` criteria, then the

R/abnormal_by_worst_grade.R

@@ -12,6 +12,8 @@
 #'   * `Any`: Numerator is number of patients with at least one abnormality, which means grade is different from 0.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("abnormal_by_worst_grade")`
+#'   to see available statistics for this function.
 #'
 #' @details The pre-processing steps are crucial when using this function. From the standard lab grade variable
 #'   `ATOXGR`, derive the following two variables:

R/abnormal_by_worst_grade_worsen.R

@@ -5,6 +5,8 @@
 #' Patient count and fraction for laboratory events (worsen from baseline) shift table.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("abnormal_by_worst_grade_worsen")`
+#'   to see all available statistics.
 #'
 #' @seealso Relevant helper functions [h_adlb_worsen()] and [h_worsen_counter()]
 #'

R/analyze_variables.R

@@ -46,6 +46,9 @@ control_summarize_vars <- control_analyze_vars
 #' functions; in this case [rtables::analyze()].
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("analyze_vars_numeric")` to see
+#'   statistics available for numeric variables, and `get_stats("analyze_vars_counts")` for statistics available
+#'   for non-numeric variables.
 #'
 #' @name analyze_variables
 NULL

R/compare_variables.R

@@ -5,6 +5,9 @@
 #' Comparison with a reference group for different `x` objects.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("analyze_vars_numeric")` to see
+#'   statistics available for numeric variables, and `get_stats("analyze_vars_counts")` for statistics available
+#'   for non-numeric variables.
 #'
 #' @note
 #' * For factor variables, `denom` for factor proportions can only be `n` since the purpose is to compare proportions

R/count_cumulative.R

@@ -7,6 +7,8 @@
 #'
 #' @inheritParams h_count_cumulative
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_cumulative")`
+#'   to see available statistics for this function.
 #'
 #' @seealso Relevant helper function [h_count_cumulative()], and descriptive function [d_count_cumulative()].
 #'

R/count_missed_doses.R

@@ -6,6 +6,8 @@
 #' mainly the special labels.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_missed_doses")`
+#'   to see available statistics for this function.
 #'
 #' @seealso Relevant description function [d_count_missed_doses()].
 #'

R/count_occurrences.R

@@ -8,6 +8,8 @@
 #' counted only once.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_occurrences")`
+#'   to see available statistics for this function.
 #'
 #' @note By default, occurrences which don't appear in a given row split are dropped from the table and
 #'   the occurrences in the table are sorted alphabetically per row split. Therefore, the corresponding layout

R/count_occurrences_by_grade.R

@@ -10,6 +10,8 @@
 #' @param grade_groups (named `list` of `character`)\cr containing groupings of grades.
 #' @param remove_single (`logical`)\cr `TRUE` to not include the elements of one-element grade groups
 #'   in the the output list; in this case only the grade groups names will be included in the output.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_occurrences_by_grade")`
+#'   to see available statistics for this function.
 #'
 #' @seealso Relevant helper function [h_append_grade_groups()].
 #'

R/count_patients_events_in_cols.R

@@ -6,6 +6,9 @@
 #' when a column table layout is required.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run
+#'   `get_stats("summarize_patients_events_in_cols")` to see available statistics for this function, in addition to any
+#'   added using `filters_list`.
 #'
 #' @name count_patients_events_in_cols
 NULL

R/count_patients_with_event.R

@@ -5,6 +5,8 @@
 #' The primary analysis variable `.var` denotes the unique patient identifier.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_patients_with_event")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [count_patients_with_flags]
 #'

R/count_patients_with_flags.R

@@ -5,6 +5,8 @@
 #' The primary analysis variable `.var` denotes the unique patient identifier.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_patients_with_flags")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [count_patients_with_event]
 #'

R/count_values.R

@@ -5,6 +5,8 @@
 #' We can count the occurrence of specific values in a variable of interest.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("count_values")`
+#'   to see available statistics for this function.
 #'
 #' @note
 #' * For `factor` variables, `s_count_values` checks whether `values` are all included in the levels of `x`

R/estimate_multinomial_rsp.R

@@ -6,6 +6,8 @@
 #' regarding the level of a factor.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("estimate_multinomial_response")`
+#'   to see available statistics for this function.
 #'
 #' @seealso Relevant description function [d_onco_rsp_label()].
 #'

R/estimate_proportion.R

@@ -5,6 +5,8 @@
 #' Estimate the proportion of responders within a studied population.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("estimate_proportion")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [h_proportions]
 #'

R/incidence_rate.R

@@ -15,6 +15,8 @@
 #'     indicating time unit for data input.
 #'   * `num_pt_year` (`numeric`)\cr time unit for desired output (in person-years).
 #' @param n_events (`integer`)\cr number of events observed.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("estimate_incidence_rate")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [control_incidence_rate()] and helper functions [h_incidence_rate].
 #'

R/odds_ratio.R

@@ -6,6 +6,8 @@
 #' along with a confidence interval.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("estimate_odds_ratio")`
+#'   to see available statistics for this function.
 #'
 #' @details This function uses either logistic regression for unstratified
 #'   analyses, or conditional logistic regression for stratified analyses.

R/prop_diff.R

@@ -3,6 +3,8 @@
 #' @description `r lifecycle::badge("stable")`
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("estimate_proportion_diff")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [d_proportion_diff()]
 #'

R/prop_diff_test.R

@@ -5,6 +5,8 @@
 #' Various tests were implemented to test the difference between two proportions.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("test_proportion_diff")`
+#'   to see available statistics for this function.
 #'
 #' @seealso [h_prop_diff_test]
 #'

R/response_biomarkers_subgroups.R

@@ -72,7 +72,7 @@ tabulate_rsp_biomarkers <- function(df,
   checkmate::assert_data_frame(df)
   checkmate::assert_character(df$biomarker)
   checkmate::assert_character(df$biomarker_label)
-  checkmate::assert_subset(vars, c("n_tot", "n_rsp", "prop", "or", "ci", "pval"))
+  checkmate::assert_subset(vars, get_stats("tabulate_rsp_biomarkers"))
 
   df_subs <- split(df, f = df$biomarker)
   tabs <- lapply(df_subs, FUN = function(df_sub) {

R/summarize_ancova.R

@@ -6,6 +6,8 @@
 #' multiple timepoints within the same response variable `.var`.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("summarize_ancova")`
+#'   to see available statistics for this function.
 #'
 #' @name summarize_ancova
 NULL

R/summarize_change.R

@@ -8,6 +8,8 @@
 #' or the change from baseline values (post-baseline) are then summarized.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("analyze_vars_numeric)`
+#'   to see available statistics for this function.
 #'
 #' @name summarize_change
 NULL

R/summarize_coxreg.R

@@ -5,6 +5,8 @@
 #' Fits a Cox regression model and estimates hazard ratio to describe the effect size in a survival analysis.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("summarize_coxreg")`
+#'   to see available statistics for this function.
 #'
 #' @details Cox models are the most commonly used methods to estimate the magnitude of
 #'   the effect in survival analysis. It assumes proportional hazards: the ratio

R/summarize_glm_count.R

@@ -6,6 +6,8 @@
 #' This can be used to analyze count and/or frequency data using a linear model.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("summarize_glm_count")`
+#'   to see available statistics for this function.
 #'
 #' @name summarize_glm_count
 NULL

R/summarize_num_patients.R

@@ -10,6 +10,8 @@
 #'   `nonunique` records.
 #' @param unique_count_suffix (`logical`)\cr should `"(n)"` suffix be added to `unique_count` labels.
 #'   Defaults to `TRUE`.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("summarize_num_patients")`
+#'   to see available statistics for this function.
 #'
 #' @name summarize_num_patients
 NULL

R/summarize_patients_exposure_in_cols.R

@@ -6,6 +6,8 @@
 #' when a column table layout is required.
 #'
 #' @inheritParams argument_convention
+#' @param .stats (`character`)\cr statistics to select for the table. Run
+#' `get_stats("analyze_patients_exposure_in_cols")` to see available statistics for this function.
 #'
 #' @name summarize_patients_exposure_in_cols
 NULL

R/survival_biomarkers_subgroups.R

@@ -213,7 +213,7 @@ tabulate_survival_biomarkers <- function(df,
   checkmate::assert_data_frame(df)
   checkmate::assert_character(df$biomarker)
   checkmate::assert_character(df$biomarker_label)
-  checkmate::assert_subset(vars, c("n_tot", "n_tot_events", "median", "hr", "ci", "pval"))
+  checkmate::assert_subset(vars, get_stats("tabulate_survival_biomarkers"))
 
   df_subs <- split(df, f = df$biomarker)
   tabs <- lapply(df_subs, FUN = function(df_sub) {

R/survival_coxph_pairwise.R

@@ -14,6 +14,8 @@
 #'   * `ties` (`string`)\cr specifying the method for tie handling. Default is `"efron"`,
 #'     can also be set to `"breslow"` or `"exact"`. See more in [survival::coxph()]
 #'   * `conf_level` (`proportion`)\cr confidence level of the interval for HR.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("coxph_pairwise")`
+#'   to see available statistics for this function.
 #'
 #' @name survival_coxph_pairwise
 NULL

R/survival_time.R

@@ -12,6 +12,8 @@
 #'   * `conf_type` (`string`)\cr confidence interval type. Options are "plain" (default), "log", or "log-log",
 #'     see more in [survival::survfit()]. Note option "none" is not supported.
 #'   * `quantiles` (`numeric`)\cr vector of length two to specify the quantiles of survival time.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("surv_time")`
+#'   to see available statistics for this function.
 #'
 #' @name survival_time
 NULL

R/survival_timepoint.R

@@ -13,6 +13,8 @@
 #'   * `conf_type` (`string`)\cr confidence interval type. Options are "plain" (default), "log", "log-log",
 #'     see more in [survival::survfit()]. Note option "none" is no longer supported.
 #'   * `time_point` (`number`)\cr survival time point of interest.
+#' @param .stats (`character`)\cr statistics to select for the table. Run `get_stats("surv_timepoint")`
+#'   to see available statistics for this function.
 #'
 #' @name survival_timepoint
 NULL