Changelog

Breaking Changes

• Adapted all modules to use teal_data objects.
• Module arguments that previously accepted inputs from teal.transform::choices_selected() or teal.transform::data_extract_spec() now only accept input from teal.transform::choices_selected(). The affected modules are: tm_a_gee, tm_a_mmrm, tm_g_ci, tm_g_forest_rsp, tm_g_forest_tte, tm_g_ipp, tm_g_km, tm_g_lineplot, tm_g_pp_adverse_events, tm_g_pp_patient_timeline, tm_g_pp_therapy, tm_g_pp_vitals, tm_t_abnormality, tm_t_abnormality_by_worst_grade, tm_t_ancova, tm_t_binary_outcome, tm_t_coxreg, tm_t_events, tm_t_events_by_grade, tm_t_events_patyear, tm_t_events_summary, tm_t_exposure, tm_t_logistic, tm_t_mult_events, tm_t_pp_basic_info, tm_t_pp_laboratory, tm_t_pp_medical_history, tm_t_pp_prior_medication, tm_t_shift_by_arm, tm_t_shift_by_arm_by_worst, tm_t_shift_by_grade, tm_t_smq, tm_t_summary, tm_t_summary_by, and tm_t_tte

Enhancements

• Updated the documentation and vignettes to demonstrate method to pass teal_data object to teal::init().
• Added parameter sort_freq_col to tm_t_events to allow the user to select column to use when sorting by decreasing frequency.
• Added parameter incl_overall_sum to tm_t_events to allow the user to choose whether overall summary rows are included at the top of the table.
• Updated the documentation and vignettes to demonstrate method to pass teal_data object to teal::init().
• Added default_total_label and set_default_total_label functions to get and set default total column label (total_label) for modules.
• Implemented tern::default_na_str and tern::set_default_na_str functions to get and set default missing value replacement string (na_level) for modules.

Bug fixes

• Fixed bug in tm_g_lineplot forcing module to initialize with a table.

Miscellaneous

• Deprecated the aval argument in tm_t_pp_laboratory and tm_g_pp_vitals and replaced it with the aval_var argument.
• Deprecated the avalu argument in tm_t_pp_laboratory and replaced it with the avalu_var argument.
• Deprecated the base_var argument in tm_g_ipp, tm_t_shift_by_arm, and template_shift_by_arm_by_worst and replaced it with the baseline_var argument.
• Specified minimal version of package dependencies.
• Replaced usage of deprecated summarize_vars function with analyze_vars.
• Reduced package dependencies (removed tidyr, rlang, magrittr and styler).

Enhancements

• Added more informative error message when grade mapping error occurs in tm_t_abnormality_by_worst_grade.
• Fixed label indentation in tm_t_abnormality_by_worst_grade.
• Added total_label argument to enable customization of the “All Patients” column/row label in the following modules: tm_a_mmrm, tm_t_abnormality, tm_t_abnormality_by_worst_grade, tm_t_binary_outcome, tm_t_events, tm_t_events_by_grade, tm_t_events_patyear, tm_t_events_summary, tm_t_exposure, tm_t_mult_events, tm_t_shift_by_arm, tm_t_shift_by_arm_worst, tm_t_shift_by_grade, tm_t_smq, tm_t_summary, tm_t_summary_by, and tm_t_tte.
• Increased default width of tm_g_forest_tte plot to prevent overlapping text.
• Improved default annotation table sizing in tm_g_km.
• Refactored tm_t_exposure to display “total” row as last row in table instead of as a summary row. Added parameters add_total_row to set whether the total row should be displayed and total_row_label to set the total row label.
• Updated tm_t_events to maintain indentation after pruning.
• Updated default reference/comparison arm level selection to work when arm variable levels are filtered out.
• Updated tm_t_coxreg to drop factor covariate variable levels that are not present to avoid errors when filtering.
• Updated tm_t_pp_basic_info, tm_t_pp_medical_history, tm_g_pp_therapy, tm_g_pp_adverse_events, and tm_t_pp_laboratory to print patient ID above table.
• Updated tm_t_pp_basic_info, tm_g_pp_therapy, tm_g_pp_adverse_events, and tm_t_pp_laboratory to use rlistings to print data neatly in reports.
• Updated tm_g_lineplot to allow user to remove interval from plot.
• Updated the documentation and vignettes to demonstrate method to pass teal_data object to teal::init().

Bug fixes

• Fixed bug in tm_t_coxreg preventing table from being displayed when no covariates are selected.

Miscellaneous

• Updated control_incidence_rate parameter names in tm_t_events_patyear from time_unit_input and time_unit_output to input_time_unit and num_pt_year, respectively, after parameter names were changed in tern.
• Hid the datasets that were not being used by the patient profile modules in the filter panel. Replaced the datanames = "all" parameter with the datasets used internally by the module.

Breaking changes

• Replaced chunks with simpler qenv class.
• Replaced datasets argument containing FilteredData with the new arguments data (tdata object) and filter_panel_api (FilterPanelAPI).

Enhancements

• Replaced synthetic_cdisc_data with refactored synthetic_cdisc_dataset function to speed up dataset loading in tests/examples.
• Added new GEE module tm_a_gee.
• Added interface for selecting an interaction term to tm_t_ancova.
• Updated encoding input checks to use shinyvalidate::InputValidator for better UI experience. Previously used shiny::validate.
• Added option to tm_a_mmrm to allow for Kenward-Roger adjustments of standard errors and p-values.
• Added option to choose facet scale options in tm_g_barchart_simple.
• Added label parameter to cs_to_select_spec/cs_to_des_select and cs_to_filter_spec/cs_to_des_filter to allow the user to customize the label printed over the selection field.
• Updated tm_t_coxreg module after refactoring summarize_coxreg in tern to fix indentation.
• Updated tm_t_exposure module to use new function analyze_patients_exposure_in_cols to fix table structure.

Bug fixes

• Fixed bug causing overlapping bars in tm_g_barchart_simple.
• Fixed bug for figures in svg format.
• Fixed bug in tm_t_summary and tm_t_summary_by preventing users from specifying the numeric_stats argument.

Miscellaneous

• Updated package Suggests to use scda.2022 rather than scda.2021.
• Removed unused argument param from tm_g_pp_vitals.
• Removed optimizer choice from tm_a_mmrm since we can always use the automatically determined optimizer.
• Created datasets to use in examples/tests for adsl, adae, adaette, adcm, adeg, adex, adlb, admh, adqs, adrs, adtte, and advs. These datasets are stored in the data folder and accessible via the tmc_ex_* prefix.
• Updated all examples and tests to use datasets from the teal.modules.clinical package instead of scda datasets.
• Updated tests to use testthat 3rd edition and replaced all applicable tests with snapshot testing.
• Implemented the lubridate package for date variables in internal data.
• Changed default value of plot_width in tm_g_forest_rsp to prevent clutter.

Enhancements

• Updated all synthetic data for tests to version rcd_2022_02_28.
• Updated all test files in tests/testthat/ to synthetic_cdisc_data("2022_02_28")
• Reverted missing data checkbox in tm_t_summary (encoding and filtering should be separate).
• Implemented a new widget that allows dragging and dropping to select comparison groups.
• Added the teal.reporter functionality to all modules.
• Enhanced the tm_t_pp_medical_history module to use the table_with_settings module and return an rtables object.
• Implemented nestcolor in examples, refactored tm_g_barchart_simple to allow use of nestcolor.
• Added more descriptive title/labels and visit name subtitle to tm_g_ci.
• Updated tm_a_mmrm column name when deselecting treatment from “all obs” to “All Patients”, added subtitles and footnotes.
• Added a title and parameter category subtitle to tm_t_exposure, cleaned up labels.
• Added titles and worse flag variable subtitles to tm_t_shift_by_grade and tm_t_shift_by_arm_by_worst.
• Added a footnote to tm_t_events_patyear with CI method.
• Added a subtitle and footnotes to tm_g_km.
• Added Stratified Analysis CI method option panel to tm_t_binary_outcome.
• Added validation for covariate/visit conflicts to tm_a_mmrm.
• Remove unnecessary brackets from header in tm_t_exposure.
• Hid footnotes in tm_g_km and tm_t_tte when “Compare Treatments” is off.

Bug fixes

• Fixed bug in tm_g_barchart_simple which prevented display of graph.
• Fixed broken example for tm_t_abnormality_by_worst_grade.
• Fixed bug in tm_a_mmrm which prevented table headers from displaying.
• Fixed bug in tm_g_forest_rsp when deselecting endpoint.
• Fixed bug in tm_t_binary_outcome that crashed the app when deselecting all paramcd.
• Fixed teal.reporter card names for tm_t_smq.
• Fixed bug in tm_t_shift_by_arm_by_worst by adding validations for choosing different endpoint values.
• Fixed bug in tm_t_coxreg preventing footnotes from displaying for univariate models.

Miscellaneous

• Added nestcolor dependency and replaced deprecated function tern::color_palette with nestcolor::color_palette.

Enhancements

Bug fixes

• Fixed summarize_logistic implementation broken by empty string error. upstream. _NA_ is the new standard flag to allow it to pivot over empty entries in data frames.
• Took out @title from tm_t_binary_outcome.R that was producing a warning.
• Updated validation to account for the error when selecting a single variable in tm_g_pp_patient_timeline module.

Miscellaneous

• Added a pkgdown template to the documentation.
• Updated package authors.

Miscellaneous

• Changed the input of Covariates in tm_t_coxreg.R to track user input and reflect the order in the table.

New features

• Added new module tm_t_shift_by_arm_by_worst for the analysis of the laboratory abnormalities with the most severe grade flag.
• Enhanced tm_t_events_patyear to include selected parameter in title of the table.
• Enhanced tm_t_mult_events to include selected parameter in title of the table.

Enhancements

• Rewrote modules to use moduleServer and updated calls to teal.devel modules which were also written to use moduleServer.
• Changed the way of obtaining of selection ordered after changes in teal.devel. Use ordered = TRUE in cs_to_des_select or cs_to_select_spec to return ordered selection.
• Replaced calls to teal::root_modules with teal::modules following deprecation of teal::root_modules.
• tm_t_events_summary now allows nested arm_var columns matching other outputs such as tm_t_events.
• Added validation in tm_t_abnormality_by_worst_grade when arm_var is not selected.
• Enhanced tm_t_binary_outcome to include all responders in the response table by default.
• Added a subtitle to tm_g_forest_tte, tm_t_coxreg, and tm_t_binary_outcome listing stratification factors.

Bug fixes

• Fixed a bug to prevent processing of empty sets of data by tm_g_forest_rsp.R causing shiny errors during the runtime.
• Fixed a bug where a closed Compare Treatments conditional panel with a marked Combine all comparison groups? option conflicted with adding a column with all patients to tables in tm_t_binary_outcome.R and tm_t_tte.R.

Miscellaneous

• Replaced the deprecated rtables::var_labels calls in the documentation examples.
• Add import of tern.mmrm package and change some references after split of tern.
• Adjusted package imports to take into account changes to the teal framework.
• Ensure consistent vertical order of tm_g_pp_patient_timeline output when switching between absolute and relative days.

Breaking changes

• Updated tm_t_abnormality due to changes in count_abnormal that abnormal argument is taking list as input now.
• Changed the tm_g_pp_patient_timeline parameter, cmtrt, to cmdecod.

New features

• Added new module tm_t_abnormality_by_worst_grade for the analysis of laboratory test results with highest grade post-baseline.
• Enhanced tm_t_ancova to include selected parameter(s), visit(s) and the analysis variable in title of the table.
• Added new module tm_g_lineplot for creating line plots.
• Enhanced tm_t_logistic to include selected parameter in title of the table.
• Enhanced tm_g_forest_rsp to include selected parameter in title of the table.
• Enhanced tm_g_forest_tte to include selected parameter in title of the table.
• Enhanced tm_g_pp_patient_timeline with bold axes labels and integer values on the axis.
• Enhanced tm_g_ipp to allow users not to display AVALU in the title nor in the y axis.

Enhancements

• Added support for logging with the logger package and added info level logs upon initialization of a module.
• Added default_responses argument to tm_t_binary_outcome and tm_g_forest_rsp to allow the user to specify default selected responses and possible response levels.
• Updated tm_t_binary_outcome to show selected responses in the output table when selecting “Show All Selected Response Categories”.
• Added rsp_table argument to tm_t_binary_outcome to allow the user to initialize the module matching the RSPT01 STREAM template.
• Added support for custom arguments for ggplot2::labs and ggplot2::theme in plot based modules.
• Added support for custom arguments for rtables::basic_table in table based modules.
• Updated tm_t_binary_outcome to enable an option to apply a continuity correction in the Wilson method.

Miscellaneous

• Updated required R version to >= 3.6.
• Refactored calls to the defunct teal.devel::data_extract_input into calls to its replacement teal.devel::data_extract_ui.
• Updated modules to use new data_merge_module interface provided by teal.devel and removed usage of the now deprecated function teal.devel::get_input_order.
• Updated tm_t_binary_outcome module to add a template and removed the now deprecated module tm_t_rsp.
• Removed utils.nest dependency and replaced calls with checkmate equivalents.

Bug Fixes

• Fixed bug in tm_g_pp_therapy where if the cmstdy or cmendy argument is of type integer causes the plot to crash.

New features

• Added new module tm_t_smq for the analysis of adverse events by Standardized MedDRA Query.
• Added new module tm_t_shift_by_grade for the analysis of grade laboratory abnormalities.
• Added new module tm_t_exposure for the analysis of duration of exposure for risk management plan.
• Added new module tm_t_shift_by_arm that can display shift table of ECG interval data.

Bug fixes

• Corrected tm_a_mmrm to be able to consider the treatment variable in all interactions.
• Fixed tm_t_binary_outcome and tm_t_rsp to choose the correct CI estimation method for Proportions Difference in Stratified Analysis (i.e. Wald-type confidence interval with CMH weights).

Enhancements

• Added validation checks to tm_t_rsp and tm_t_binary_outcome for stratification errors from applied filters.
• Added tm_g_km validation check for plot tables font size.
• Enhanced tm_g_km to add the selected paramcd in the plot title.
• tm_t_events now can display layouts with two nested column as treatment variables. The same options for pruning and sorting are available.
• Exported package helper functions.
• Updated tm_t_events_by_grade to display grading groups in nested columns with col_by_grade option and support pruning and sorting options like tm_t_events.
• Used format_count_fraction to fix formatting inconsistency in tm_t_events_summary.
• Updated count_occurrences vars argument in tm_t_shift_by_grade.
• Updated tm_t_pp_laboratory to display 4 decimals by default.
• Updated tm_t_events_by_grade to use the trim_levels_in_group split function instead of the trim_rows function.
• Added a table title to tm_t_tte.
• Added table titles to tm_t_rsp and tm_t_binary_outcome.

Miscellaneous

• Updated LICENCE and README with new package references.
• Added error_on_lint: TRUE to .lintr.
• Removed insert_rrow and updated usage of count_patients_by_flags in tm_t_events_summary.
• Changed how the package calls functions from the dplyr package. The functions should be now fully specified (e.g. dplyr::filter).

New features

• Added capability to remember the order of user input to some encoding UI elements. Inputs marked with a double arrow icon have tracking enabled. The affected modules are: tm_t_summary, tm_t_summary_by, tm_g_forest_rsp, tm_g_forest_tte, tm_t_events_summary, tm_t_abnormality, tm_t_mult_events.
• Added a new argument numeric_stats to tm_t_summary and tm_t_summary_by to control displayed summary statistics for numeric variables.
• Added a new argument drop_zero_levels to tm_t_summary_by so that you can drop rows with all zeros from result table.

Enhancements

• Split tm_g_patient_profile tabs into 8 separate new modules.
• Added the option to select patient ID from the filter panel for all the modules of patient profile.
• Added a validation for tm_g_patient_timeline when the plot is empty.
• Enhanced tm_a_mmrm to work without the treatment variable.
• Added the option to choose the number of decimal places for rounding in tm_t_pp_laboratory.
• Added a check box to tm_g_pp_patient_timeline hiding/showing relative study days on the x-axis.
• Added a title with patient’s id to plots in patient profile modules.
• Made the gray error message in tm_g_forest_tte more informative when deselecting the Endpoint column in the left-hand encoding panel.
• Added the twenty-fifth and seventy-fifth quantile to summary statistics in tm_t_summary.
• Added an interaction p-value column for tm_t_coxreg.
• Added a validation for tm_t_ancova when selected covariate variables contain one level.
• Added a validation for tm_t_events_patyear when the events variable is empty.
• Changed the font size input description for tm_g_km to more precisely describe what it controls.
• Enhanced tm_t_logistic so that interaction choices depend on the selected covariates.
• Enhanced tm_t_rsp so that strata input is visible when comparing treatments.

Bug fixes

• Fixed Get R Code output of tm_t_pp_laboratory to return identical HTML formatted table as displayed in the app.
• Added a validation for tm_t_coxreg to ensure treatment, strata and covariate variables do not overlap.
• Limited the label repel feature in tm_g_pp_patient_timeline to X-axis for a more consistent look.
• Updated tm_t_summary_by so that paramcd is not required when analyzing ADSL variables.
• Updated tm_t_coxreg so that it can work when there is no covariate selected.
• Updated tm_a_mmrm so that it can work when treatment variable is not selected.
• Updated tm_g_forest_tte so that total number of events are also shown in the table.
• Updated tm_t_events_summary to work with pooled studies.
• Updated validation for the at level of tm_t_coxreg.
• Updated validation for the at level of tm_t_logistic.
• Added a validation for tm_t_binary_outcome and tm_t_rsp to ensure strata variable contains more than one level when selecting one strata variable.
• Updated the warning message when deselecting all statistics in tm_t_summary and tm_t_summary_by to explain need to select at least one statistic.

Enhancements

• Added the option to download and expand tables.
• In tm_g_km added support for downloading images and updated x-axis label to show in title case.
• For tm_g_patient_profile:
  • Added a slider for the font size in plots.
  • Added persistence for selected table lengths.
  • The timeline plot now supports more edge cases.
  • In vitals tab, removed unused label text legend, updated plot to display stable colors per levels, cleared x-axis limit and fixed legend to update when filtering. Also added a note to clarify the supported horizontal lines cases.
  • Updated adverse events tab to show a warning message instead of an empty plot when data is empty.
  • Fixed PARAMCD selected levels for current patient.
• For tm_t_tte:
  • Updated the pre-processing code inside template_tte so that a dataset without any events still produces a table.
  • Updated code to use correct denominator for duration of response endpoints.
• For tm_t_summary:
  • Modified the parameter arm_var to accept more than one column. When selecting two columns for arm_var, the second variable will nest under the first one.
  • Added argument show_labels to template_summary to show the label for a single summary variable in the table.
• Added a new parameter conf_arg to tm_t_rsp to be consistent with other efficacy modules.
• Added validation statement in tm_g_ipp module to print message when deselecting Timepoint Variable drop-down.
• Removed header definition in tm_g_forest_rsp and tm_g_forest_tte as there is now a default header in g_forest.
• Fixed validation statement in tm_t_coxreg so that models without strata and using likelihood tests return a result.
• Clarified functionality of drop_arm_levels for tm_t_summary and tm_t_summary_by. In the encodings panel, the checkbox will show when the parent dataset and analysis dataset are different.

Miscellaneous

• Replaced the remaining two observe function calls with observeEvent to optimize performance.
• Fixed grammar in the “Select a patient’s id” error message in the tm_g_patient_profile.
• Fixed the font_size default of the templates to be 12L instead of a vector of 3 integers and cleaned associated unnecessary code.
• Fixed deprecated function warning in tm_g_barchart_simple.
• Fixed subgroup_var definition truncation in tm_g_forest_rsp and tm_g_forest_tte.
• Clarified labeling related to regression type in the encoding panel from tm_t_coxreg.

Bug fixes

• Added a validation for the case when filtering out all rows in the therapy tab of tm_g_patient_profile.
• Updated the internals of the modules to read data from the correct field of the filter_spec objects.
• Fixed the reactivity between the filter panel and the PARAMCD variable levels input in tm_g_patient_profile vitals tab so that the plot does not get reset when filtering.
• Updated the vitals plot tab in tm_g_patient_profile to drop NA entries in the plot.
• Updated tm_t_coxreg to take at values into account.
• Added a check in tm_t_coxreg to have interactions in univariate models but not in multivariate models.
• Updated tm_t_events_summary to work with pooled studies.

New Module

• Added new module tm_g_patient_profile to profile patients based on predefined categories.
• Added new module tm_g_ipp for individual patient plots.

Enhancements

• Added the argument drop_arm_levels to all safety modules. This allows removal of columns based on factor levels not found in filtered data.
• Updated tm_g_km to allow plot of failure probability on y-axis, tick interval selection on x-axis and option to create plot without confidence interval ribbon (new default).
• Added the argument time_unit_var to template_g_km to add the time unit to the x-axis label.

Miscellaneous

• Removed redundant Analysis Data: label from Encodings Panel.
• Removed limit requiring 15 or fewer columns for tabulation modules. New upper threshold is 100 columns.
• Decreased the lower limit for number of observations required by modules. Safety tables require at least one record. Requirements for efficacy outputs per treatment group: tm_a_mmrm requires five records, tm_t_logistic and tm_t_coxreg require two records, and the remaining modules require at least one record per treatment group. For graphs, the lower threshold is two records.
• Removed argument cnsr_val from tm_t_events_patyear and added new argument events_var.
• arm_ref_comp_observer to include parentname argument.
• Show R code to include datasets retrieved from data_extract_spec objects.
• Refactored out the stringr dependency from the patient profile module.
• Added missing table calls in chunks for tm_t_events and tm_t_events_by_grade.

New Features

• Added new module tm_g_ci for confidence interval plots.
• Added new module tm_t_ancova for analysis of variance summary tables.
• Added new module tm_t_mult_events for multi-event tables.

Enhancements

• Refactored all modules using the redesigned rtables and tern packages.
• Enhanced modules. They now take advantage of data_extract_spec and data_merge_module functionality from teal.
• Reduced clutter from repeated datasets in the encodings panels.
• Updated all modules to use OptionalSelectInput for conf_level.

Miscellaneous

• Added vignette about substitute which can be helpful when developing analysis template functions for teal modules.

Bug fixes

• Updated tm_t_events module to use user’s non-default choices for prune_freq and prune_diff.