v0.9.10

Enhancements

• Added alternative argument to test_proportion_diff() to allow one-sided hypothesis testing.
• Added cmh_sato method to estimate_proportion_diff() for Cochran-Mantel-Haenszel proportion difference confidence interval using the Sato variance estimator.
• Added cmh_mn method to estimate_proportion_diff() for Cochran-Mantel-Haenszel proportion difference confidence interval using the Miettinen and Nurminen method.
• Added cmh_wh (CMH with Wilson-Hilferty transformation) method to test_proportion_diff() for stratified proportion difference testing.
• analyze_vars now accepts format, formats_var and na_strs_var and passes them directly down to analyze
• When .formats is "default" (the string), all stats will be assigned the "default" format label rather than the default formats baked into tern. This is intended for use in conjunction with the format or formats_var arguments to analyze (and now analyze_vars).

Bug Fixes

• Fixed bug in tabulate_rsp_subgroups() and tabulate_survival_subgroups() preventing risk difference column format specified via control_riskdiff() from being applied.
• Fixed bug in a_summary() causing an error when all values of a factor input variable were NA.
• Fixed NA behavior in s_summary.factor and s_summary.character when na.rm = FALSE to ensure that NA values are always represented in output tables.
• Unified explicit NA behavior for missing values in explicit_na().

What's Changed

• Update .revdeprefs.yaml by @shajoezhu in #1430
• Fix bug preventing control_riskdiff(format) from being recognized by @edelarua in #1434
• update "matches" code in g_lineplot to retain level order. by @ayogasekaram in #1437
• Fix a_summary() all-NA factor bug by @iaugusty in #1432
• Fix NAs behavior by @Melkiades in #1441
• Add lme4 as a suggested dependency by @shajoezhu in #1443
• 1444: Added alternative argument to test_proportion_diff() by @danielinteractive in #1446
• Issue1445 internal branch by @gmbecker in #1452
• 1444/1445: Two more CMH prop diff (test) methods by @danielinteractive in #1454
• hook up support for "default" format and formats_var by @gmbecker in #1456
• Bump version to 0.9.10 and clean up DESCRIPTION [skip vbump] by @shajoezhu in #1458

Full Changelog: v0.9.9...v0.9.10

v0.9.9

Enhancements

• Updated tabulate_rsp_subgroups() and tabulate_survival_subgroups() to specify parameter parent_name when using split_rows_by() and analyze() internally to enhance table paths.

Bug Fixes

• Fixed bug in g_lineplot() where table_format and table_labels arguments were ignored.
• Fixed bug in g_lineplot() to allow use of a function as a format for table stats.

Miscellaneous

• Converted assert_df_with_factors(), assert_df_with_variables(), assert_proportion_value(), check_diff_prop_ci(), clogit_with_tryCatch(), get_covariates(), labels_or_names(), range_noinf() , s_surv_timepoint(), and s_test_proportion_diff() to exported functions.

What's Changed

• Fix warning from get_formats_from_stats() by @edelarua in #1420
• 1418 [Bug]: g_lineplot doesn't work with a format as a function by @yurovska in #1421
• alter var_labels by @shajoezhu in #1423
• 1425 feature request export various internal functions for usage in jj based sister package to nest tern by @iaugusty in #1426
• v0.9.9 Release [skip vbump] by @edelarua in #1427

Full Changelog: v0.9.8...v0.9.9

v0.9.8.433

Enhancements

• Refactored count_abnormal(), count_abnormal_by_baseline(), count_abnormal_by_marked(), count_abnormal_by_worst_grade(), count_abnormal_lab_worsen_by_baseline(), count_cumulative(), count_missed_doses(), coxph_pairwise(), estimate_multinomial_rsp(), estimate_proportion(), estimate_proportion_diff(), estimate_odds_ratio(), summarize_ancova(), summarize_glm_count(), summarize_num_patients(), surv_timepoint(), and test_proportion_diff() to work without make_afun().
• Refactored afun_riskdiff(), count_occurrences(), count_occurrences_by_grade(), count_patients_with_event(), count_patients_with_flags(), count_values(), estimate_incidence_rate(), h_tab_one_biomarker(), summarize_change(), summarize_colvars(), summarize_patients_exposure_in_cols(), survival_time(), tabulate_rsp_subgroups(), tabulate_survival_subgroups(), tabulate_rsp_biomarkers(), and tabulate_survival_biomarkers() to align with new analysis function style.
• Added geom_sd and geom_mean_sd to s_summary() as default available statistics.
• Added denom parameter to estimate_proportion().
• Added weights_emmeans argument to summarize_ancova().

Bug Fixes

• Fixed bug in a_count_patients_with_flags() preventing select custom label/indentation formats from being applied.
• Fixed bug in tabulate_rsp_subgroups() and tabulate_survival_subgroups() preventing the pct option from having an effect when adding a risk difference column.
• Fixed bug with the order of .stats when adding custom statistical functions.
• Fixed bug with multiple custom functions not being represented correctly as a list of output stats.

Miscellaneous

• Began deprecation of the unused table_names argument to count_abnormal_lab_worsen_by_baseline().
• Began deprecation of the unused h_split_param() function.
• Began deprecation of the unused label_all parameter to tabulate_rsp_subgroups(), with redirection to the same parameter in extract_rsp_subgroups().
• Began deprecation of the no longer used helper functions h_tab_one_biomarker(), h_tab_rsp_one_biomarker(), and h_tab_surv_one_biomarker().
• Updated documentation to remove suggestions to use make_afun().
• Added warnings for geom_mean statistical output.
• Moved helper functions h_tab_rsp_one_biomarker() and h_tab_surv_one_biomarker() into h_biomarkers_subgroups.R.
• Reorganized the utility documentation related to factors (utils_factor.R) into a single file.
• Converted as_factor_keep_attributes() to an exported function.
• Removed internal function ungroup_stats() and replaced its usage with the get_*_from_stats() functions.
• Removed s_count_nonmissing() as it is a non-repeated small and internal function.

Full Changelog: v0.9.8...v0.9.8.433

v0.9.8

Enhancements

• Refactored count_abnormal(), count_abnormal_by_baseline(), count_abnormal_by_marked(), count_abnormal_by_worst_grade(), count_abnormal_lab_worsen_by_baseline(), count_cumulative(), count_missed_doses(), coxph_pairwise(), estimate_multinomial_rsp(), estimate_proportion(), estimate_proportion_diff(), estimate_odds_ratio(), summarize_ancova(), summarize_glm_count(), summarize_num_patients(), surv_timepoint(), and test_proportion_diff() to work without make_afun().
• Refactored afun_riskdiff(), count_occurrences(), count_occurrences_by_grade(), count_patients_with_event(), count_patients_with_flags(), count_values(), estimate_incidence_rate(), h_tab_one_biomarker(), summarize_change(), summarize_colvars(), summarize_patients_exposure_in_cols(), survival_time(), tabulate_rsp_subgroups(), tabulate_survival_subgroups(), tabulate_rsp_biomarkers(), and tabulate_survival_biomarkers() to align with new analysis function style.
• Added geom_sd and geom_mean_sd to s_summary() as default available statistics.
• Added denom parameter to estimate_proportion().
• Added weights_emmeans argument to summarize_ancova().

Bug Fixes

• Fixed bug in a_count_patients_with_flags() preventing select custom label/indentation formats from being applied.
• Fixed bug in tabulate_rsp_subgroups() and tabulate_survival_subgroups() preventing the pct option from having an effect when adding a risk difference column.
• Fixed bug with the order of .stats when adding custom statistical functions.
• Fixed bug with multiple custom functions not being represented correctly as a list of output stats.

Miscellaneous

• Began deprecation of the unused table_names argument to count_abnormal_lab_worsen_by_baseline().
• Began deprecation of the unused h_split_param() function.
• Began deprecation of the unused label_all parameter to tabulate_rsp_subgroups(), with redirection to the same parameter in extract_rsp_subgroups().
• Began deprecation of the no longer used helper functions h_tab_one_biomarker(), h_tab_rsp_one_biomarker(), and h_tab_surv_one_biomarker().
• Updated documentation to remove suggestions to use make_afun().
• Added warnings for geom_mean statistical output.
• Moved helper functions h_tab_rsp_one_biomarker() and h_tab_surv_one_biomarker() into h_biomarkers_subgroups.R.
• Reorganized the utility documentation related to factors (utils_factor.R) into a single file.
• Converted as_factor_keep_attributes() to an exported function.
• Removed internal function ungroup_stats() and replaced its usage with the get_*_from_stats() functions.
• Removed s_count_nonmissing() as it is a non-repeated small and internal function.

What's Changed

• Fix label and indentation processors by @edelarua in #1379
• Remove the ungroup_stats() function by @edelarua in #1385
• Fix pct bug in tabulate_rsp_subgroups when risk difference column is added by @edelarua in #1387
• Retire make_afun() - Batch 3 by @edelarua in #1389
• Remove lookup-refs by @m7pr in #1395
• Fix failed scheduled test by @shajoezhu in #1396
• refactor geom sd and fix custom stat order by @Melkiades in #1393
• Retire make_afun() - Batch 2 by @edelarua in #1392
• Add in missing analysis function option for custom statistics by @edelarua in #1400
• Retire make_afun() - Batch 4 by @edelarua in #1394
• Refactor tabulate_*() functions and deprecate their helper functions by @edelarua in #1398
• Refactor remaining analysis functions by @edelarua in #1401
• Batch 1 by @Melkiades in #1403
• Fix analysis function refactor bugs by @edelarua in #1406
• upversion r min version, and survial depednecies by @shajoezhu in #1408
• feature: denom addition to estimate_proportions() by @Melkiades in #1404
• Update .revdeprefs.yaml by @shajoezhu in #1410
• [821] Feature Request: Add weights option to ancova by @yurovska in #1412
• v0.9.8 Release [skip vbump] by @edelarua in #1413

New Contributors

• @yurovska made their first contribution in #1412

Full Changelog: v0.9.7...v0.9.8

v0.9.7 cran

Enhancements

• Added the denom parameter to s_count_cumulative(), s_count_missed_doses(), and s_count_occurrences_by_grade().
• Added "N_row" as an optional input to denom in s_count_occurrences().
• Added rel_height_plot parameter to g_lineplot() to control the line plot height relative to annotation table height.
• Added as_list parameter to g_lineplot() to allow users to return the line plot and annotation table elements as a list instead of stacked for more complex customization.
• Added vignette "Understanding tern functions" for future reference.
• Added to analyze_vars() statistical names that are used by rtables::as_result_df().
• Added the possibility to integrate custom statistical functions to default ones in analyze_vars().
• Refactored analyze_vars() and a_summary() to take all options from ?rtables::additional_fun_params.
• Refactored summarize_change() and count_values() to work without make_afun().
• Refactored a_count_occurrences_by_grade(), a_count_patients_with_event(), and a_count_patients_with_flags() to no longer use make_afun().
• Refactored get_labels_from_stats() to use a named list of levels for each statistic instead of row names.
• Updated the table_font_size parameter of g_lineplot() to control the size of all text in the annotation table, including labels.
• Merged compare_vars() into analyze_vars() as overlap was significant.

Bug Fixes

• Fixed bug in a_summary() causing non-unique row_name values to occur when multiple statistics are selected for count variables.

Miscellaneous

• Reverted deprecation of quick get functions summary_formats() and summary_labels(). Added disclaimer about underlying use of get_stats.
• Corrected handling of extra arguments and NA for summarize_change().
• Removed count_fraction_fixed_dp exception by assigning it to the result of count_fraction with a different format output.

What's Changed

• 1315 var labels odds ratio@main by @ayogasekaram in #1317
• Update NEST site URLs by @edelarua in #1319
• Add method argument to s_odds_ratio() and estimate_odds_ratio() by @edelarua in #1320
• Update DESCRIPTION by @shajoezhu in #1321
• Fix bug in a_summary() causing nonunique row_name values by @edelarua in #1323
• Pkgdown site improvements by @edelarua in #1328
• List options in .stats argument descriptions by @edelarua in #1330
• Adds denom argument to count s_* functions by @edelarua in #1326
• Move nestcolor to imports by @edelarua in #1339
• import nestcolor by @shajoezhu in #1341
• close #1340, rm staged dep by @shajoezhu in #1344
• Add table formatting options to g_lineplot() - table relative height & font size by @edelarua in #1335
• Refactor count_patients_with_event() and count_patients_with_flags() by @edelarua in #1343
• Fix indent modifier bug in count_patients_with_flags() by @edelarua in #1350
• Refactor summarize_change() by @Melkiades in #1347
• 1301 feature request add confidence intervals for quantiles in surv time by @iaugusty in #1306
• adding offset to nb by @shajoezhu in #1358
• Test coxph by @shajoezhu in #1365
• Refactoring of analyze_vars with specifics for .stat_names by @Melkiades in #1361
• Refactor a_count_values() by @edelarua in #1374
• vbump to 0.9.7, prep for release [skip vbump] by @shajoezhu in #1372

New Contributors

• @iaugusty made their first contribution in #1306

Full Changelog: v0.9.6...v0.9.7

v0.9.6 cran

Enhancements

• Added errorbar_width and linetype parameters to g_lineplot.
• Added the .formats argument to tabulate_rsp_subgroups and tabulate_survival_subgroups to allow users to specify formats.
• Added the riskdiff argument to tabulate_rsp_subgroups and tabulate_survival_subgroups to allow users to add a risk difference table column, and function control_riskdiff to specify settings for the risk difference column.
• Added warning to tabulate_rsp_subgroups when pval statistic is selected but df has not been correctly generated to add p-values to the output table.
• Added n_rate statistic as a non-default option to estimate_incidence_rate which returns both number of events observed and estimated incidence rate.
• Added n_unique statistic as a non-default option to estimate_incidence_rate which returns total number of patients with at least one event observed.
• Refactored estimate_incidence_rate to work as both an analyze function and a summarize function, controlled by the added summarize parameter. When summarize = TRUE, labels can be fine-tuned via the new label_fmt argument to the same function.
• Added fraction statistic to the analyze_var_count method group.
• Improved summarize_glm_count() documentation and all its associated functions to better describe the results and the functions' purpose.

Bug Fixes

• Added defaults for d_count_cumulative parameters as described in the documentation.
• Fixed a bug of empty levels on g_lineplot x-axis were not shown in either plots.
• Fixed a bug in a_surv_time that threw an error when split only has "is_event".
• Fixed disappearing line in g_lineplot when using only one group or strata level.
• Fixed defaults for formats and labels in get_formats_from_stats and get_labels_from_stats.
• Fixed bug for linear scaling factor (scale parameter) being applied to response but not to rate in h_glm_count while all distributions have logarithmic link function.
• Fixed bug in decorate_grob that did not handle well empty strings or NULL values for title and footers.
• Fixed bug in g_km that caused an error when multiple records in the data had estimates at max time.
• Fixed issue with wrong wrapping due to different \n and vector behavior that did not cope well with split_string().

Miscellaneous

• Began deprecation of the confusing functions summary_formats and summary_labels.
• Enhanced general descriptions of analyze and summarize functions throughout package documentation.
• Finalized deprecation of the strata and cohort_id arguments to g_lineplot.
• Moved incidence rate helper functions into a separate h_incidence_rate.R file.

What's Changed

• Bumps {rmarkdown} minimal version by @averissimo in #1247
• Fixes problem with a_surv_time with datasets where "is_event" is always TRUE by @averissimo in #1260
• Update of g_lineplot to have line type, error bar width, and avoid empty levels in by @Melkiades in #1261
• Fixing parameters' defaults for d_count_cumulative by @Melkiades in #1264
• Fix typo in has_fraction_in_any_col example by @edelarua in #1266
• update feature yml by @shajoezhu in #1270
• rework docs summarize_glm_count by @Melkiades in #1269
• Fix get_labels_from_stats test by @edelarua in #1272
• Workflow propagations by @walkowif in #1275
• add lookup-refs param to rhub workflow by @pawelru in #1277
• Add riskdiff argument to tabulate_rsp_subgroups and tabulate_survival_subgroups by @edelarua in #1276
• fix decorate_grob for empty strings by @Melkiades in #1278
• Fix g_km bug when multiple estimates are made at max time by @edelarua in #1280
• Update DESCRIPTION by @pawelru in #1288
• Update DESCRIPTION emmeans to 1.10.4 by @shajoezhu in #1284
• update readme by @pawelru in #1290
• Fix URLs by @edelarua in #1298
• Add warning for missing p-value in tabulate_rsp_subgroups() output by @edelarua in #1296
• Add n_rate statistic to estimate_incidence_rate() by @edelarua in #1295
• Improve analyze/summarize function documentation by @edelarua in #1283
• Finalize deprecation of <0.9.4 arguments by @edelarua in #1304
• fix grob wrapping final by @Melkiades in #1305
• Workflow propagations by @walkowif in #1291
• Update DESCRIPTION, upversion formatters minimal version to 0.5.9 by @shajoezhu in #1308
• feat: added "fraction" statistic to the "analyze_var_count" method group by @kpagacz in #1309
• resolve green ci by @shajoezhu in #1293
• Refactor estimate_incidence_rate by @edelarua in #1300
• prep for rspm release 0.9.6, [skip vbump] by @shajoezhu in #1311

Full Changelog: v0.9.5...v0.9.6

v0.9.5 cran

Enhancements

• Added facet_var to g_lineplot to allow plot faceting by a factor variable.
• Added examples and tests for label_all parameter to extract_survival_biomarkers and extract_survival_subgroups.
• Added the xticks, xlim, and ylim arguments to g_lineplot to allow for customization of the x and y axes.
• Updated g_lineplot legend to follow factor levels set by users.

Bug Fixes

• Fixed a bug in s_ancova that prevented statistics from being printed when arm levels include special characters.
• Fixed a bug in decorate_grob that prevented the right margins to be respected when adding title and footers decorations.

Miscellaneous

• Began deprecation of the unused label_all parameter to tabulate_survival_biomarkers and tabulate_survival_subgroups, with redirection to the same parameter in their associated extract_* functions.

What's Changed

• Fix documentation for the label_all argument of extract_* functions by @edelarua in #1232
• Manually specify shapes within g_lineplot when >6 group levels exist by @edelarua in #1234
• update g_lineplot code to handle special characters by @ayogasekaram in #1230
• Add revdepcheck configuration by @walkowif in #1228
• Fix bug in s_ancova when arm levels include special characters by @edelarua in #1242
• Fix indentation in functions that use matrix_form by @edelarua in #1239
• Add xticks, xlim, and ylim arguments to g_lineplot by @edelarua in #1244
• fix decorate grob vp widths for titles and footers when right margin is present by @Melkiades in #1245
• Truetype pagination PR check by @gmbecker in #1250
• testing for rtables higher level ns PR by @gmbecker in #1251
• prep for 0.9.5 [skip vbump] by @shajoezhu in #1253

New Contributors

• @gmbecker made their first contribution in #1250

Full Changelog: v0.9.4...v0.9.5

v0.9.4

New Features

• Added h_glm_negbin to h_glm_count to enable count data analysis using a negative binomial model.
• Added argument grade_groups_only to count_occurrences_by_grade to allow users to only display rows for specified grade groups.
• Added internal function df2gg that converts data.frame objects to ggplot objects.
• Added functions control_surv_med_annot and control_coxph_annot to configure g_km annotation table sizes/positions.
• Refactored g_km to output a ggplot object instead of a grob object.
• Refactored g_forest to output a ggplot object instead of a grob object.

Enhancements

• Default format for mean p-values (mean_pval) updated from "xx.xx" to "x.xxxx | (<0.0001)".

Bug Fixes

• Fixed nested column split label overlay issue in rtable2gg to clean up appearance of text labels.
• Fixed bug in s_ancova causing incorrect difference calculations for arm variables with irregular levels.
• Fixed bug in format_count_fraction_fixed_dp that did not have the same print when the fraction was 1 (100%).
• Fixed bug in g_lineplot causing default labels not to update according to specified control settings.
• Fixed bug causing an error when automatic value formatting was applied to NA values.

Miscellaneous

• Added function expect_snapshot_ggplot to test setup file to process plot snapshot tests and allow plot dimensions to be set.
• Adapted to argument renames introduced in ggplot2 3.5.0.
• Renamed individual_patient_plot.R to g_ipp.R.
• Removed all instances of deprecated parameters time_unit_input, time_unit_output, na_level and indent_mod.
• Removed deprecated functions summarize_vars, control_summarize_vars, a_compare, create_afun_summary, create_afun_compare, and summary_custom.
• Removed vdiffr package from Suggests in DESCRIPTION file.
• Began deprecation of the named element strat, to be renamed to strata, within the variables argument to h_rsp_to_logistic_variables, h_logistic_mult_cont_df, h_odds_ratio_subgroups_df, h_coxreg_mult_cont_df, h_coxph_subgroups_df, h_tbl_coxph_pairwise, extract_rsp_biomarkers, extract_rsp_subgroups, extract_survival_biomarkers, and extract_survival_subgroups.
• Began deprecation of the strat argument to s_coxph_pairwise and replaced it with the strata argument.
• Began deprecation of the forest_grob, forest_dot_line, forest_viewport, vp_forest_table_part, and grid.forest functions.
• Began deprecation of no longer used Kaplan-Meier helper functions h_ggkm, h_decompose_gg, h_km_layout, h_grob_tbl_at_risk, h_grob_median_surv, h_grob_y_annot, and h_grob_coxph.
• Began deprecation of grob/grid related functions stack_grobs, arrange_grobs, and draw_grob which are no longer used in tern.

What's Changed

• Workflow propagation by @walkowif in #1159
• Refactor g_forest by @edelarua in #1158
• Fix graph snapshot tests by @edelarua in #1165
• update snapshots by @ayogasekaram in #1166
• add h_glm_negbin() by @duanx9 in #1163
• Fix irregular arm levels calculation bug in s_ancova by @edelarua in #1169
• Add option grade_groups_only to count_occurrences_by_grade by @edelarua in #1170
• Fix README links by @edelarua in #1174
• Adding Bland- Altman statistical function by @linc50 in #1135
• Fix warnings in examples by @edelarua in #1180
• fix roxygenise deps; styler by @pawelru in #1183
• Fix pkgdown redirect by @vedhav in #1185
• Propagate files from insightsengineering/.github by @insights-engineering-bot in #1187
• [skip actions] Propagate files from insightsengineering/.github by @insights-engineering-bot in #1188
• #nolintr to # nolint by @pawelru in #1189
• Update DESCRIPTION, update author list by @shajoezhu in #1190
• Skip plot snapshot testing by @edelarua in #1194
• Fix typo, ggplot2 update warnings by @edelarua in #1197
• Replace strat with strata by @edelarua in #1198
• fix format precision for format_count_fraction_fixed_dp by @Melkiades in #1192
• options for strict tests; few enhancements by @pawelru in #1182
• Fix integration tests by @edelarua in #1200
• Update default format for mean p-values by @edelarua in #1204
• rename trans to transform by @pawelru in #1202
• add snaps for bland-altman function by @ayogasekaram in #1201
• Remove deprecated functions, deprecate forest plot helpers by @edelarua in #1211
• Fix g_lineplot default labels by @edelarua in #1213
• Refactor g_km by @edelarua in #1210
• Accept NULL to get_labels_from_stats by @edelarua in #1215
• Revert deprecation of decorate_grob-related functions used by citril by @edelarua in #1219
• fix verdepcheck by @pawelru in #1216
• Fix auto formatting for NA values by @edelarua in #1220
• Minor documentation clean up by @edelarua in #1222
• Workflow propagation by @walkowif in #1206
• add faceting variable to g_lineplot. add code to maintain factor levels. by @ayogasekaram in #1226
• Add tests for functions with missing coverage by @edelarua in #1229
• [skip vbump], preping for 0.9.4 release by @shajoezhu in #1227

New Contributors

• @duanx9 made their first contribution in #1163
• @linc50 made their first contribution in #1135
• @vedhav made their first contribution in #1185

Full Changelog: v0.9.3...v0.9.4

v0.9.3

New Features

• Added ref_group_position function to place the reference group facet last, first or at a certain position.
• Added keep_level_order split function to retain original order of levels in a split.
• Added level_order split function to reorder manually the levels.
• Added function get_indents_from_stats to format and return indent modifiers for a given set of statistics.
• 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.
• Added summarize function version of count_occurrences analyze function, summarize_occurrences.
• Added referential footnotes to surv_time for censored range observations, controlled via the ref_fn_censor parameter.
• Added helper function h_adlb_abnormal_by_worst_grade to prepare ADLB data to use as input in count_abnormal_by_worst_grade.
• Added function rtable2gg that converts rtable objects to ggplot objects.
• Added helper function to set default na_str globally with set_default_na_str() and added default_na_str() for all interested functions.

Enhancements

• Added ref_group_coxph parameter to g_km to specify the reference group used for pairwise Cox-PH calculations when annot_coxph = TRUE.
• Added annot_coxph_ref_lbls parameter to g_km to enable printing the reference group in table labels when annot_coxph = TRUE.
• Added x_lab parameter to g_lineplot to customize x-axis label.
• Remove 25% padding of y-axis in g_lineplot.
• Added support for creating multiple risk difference columns, each comparing to a single comparison group. Multiple comparison groups can be specified as a vector via the arm_y argument.
• Allowed numeric vector as count_by input in analyze_num_patients and summarize_num_patients.
• Aligned plot and table vertically in g_lineplot.

Bug Fixes

• Fixed bug in decorate_grob preventing text wrapping from accounting for font size.
• Fixed implementation of na_str argument in all column-wise analysis and tabulation functions.

Miscellaneous

• Specified minimal version of package dependencies.
• Upgraded to_string_matrix to take into account widths and other printing parameters.

What's Changed

• Adds min package version by @averissimo in #1012
• Improve documentation of whiskers param in g_lineplot by @edelarua in #1110
• Refactor count_occurrences by @edelarua in #1076
• Add split functions for ref_group position by @Melkiades in #1111
• change cohort_id and strata to subject_var and group_var. by @ayogasekaram in #1106
• Fix integrations tests after ref_group update by @edelarua in #1115
• Add option to set reference arm in g_km by @edelarua in #1120
• Add x_lab parameter to g_lineplot by @edelarua in #1126
• Fix decorate_grob text wrapping by @edelarua in #1124
• Add utility functions for labels and formats by @edelarua in #1122
• Remove extra y-axis padding in g_lineplot by @edelarua in #1128
• Expand get_stats to cover all analysis functions by @edelarua in #1123
• Refactor surv_time to enable cell footnotes by @edelarua in #1116
• Improve analyze_vars documentation by @edelarua in #1129
• Add helper function to pre-process ADLB for count_abnormal_by_worst_grade by @edelarua in #1132
• Reorder doc pages so analyze functions appear first, add links in README by @edelarua in #1131
• Expose statistic function arguments in analyze functions by @edelarua in #1134
• Implement na_str argument in all columnwise analysis functions by @edelarua in #1140
• Add support for multiple risk difference columns by @edelarua in #1143
• Allow numeric input to count_by in analyze_num_patients by @edelarua in #1141
• Add function to convert rtables to ggplots by @edelarua in #1137
• Add "experimental" badge to rtable2gg by @edelarua in #1147
• Generate pkgdown site favicon by @edelarua in #1149
• Add svglite to Suggests by @edelarua in #1151
• Vertically align plot with table in g_lineplot by @edelarua in #1153
• Fix ggsave tests by @edelarua in #1154
• Add general default for na_str by @Melkiades in #1145
• add axis parameter to plot_grid() by @pawelru in #1155
• update g_lineplot snaps by @pawelru in #1157
• init commit for preping new cran release [skip vbump] by @shajoezhu in #1156

New Contributors

• @averissimo made their first contribution in #1012

Full Changelog: v0.9.2...v0.9.3

v0.9.2

New Features

• Added the na_str argument to analyze & summarize_row_groups wrapper functions count_abnormal, count_abnormal_by_baseline, count_abnormal_by_marked, count_abnormal_by_worst_grade, count_abnormal_lab_worsen_by_baseline, count_cumulative, count_missed_doses, count_occurrences, count_occurrences_by_grade, summarize_occurrences_by_grade, summarize_patients_events_in_cols, count_patients_with_event, count_patients_with_flags, count_values, estimate_multinomial_response, estimate_proportion, h_tab_one_biomarker, estimate_incidence_rate, logistic_summary_by_flag, estimate_odds_ratio, estimate_proportion_diff, test_proportion_diff, summarize_ancova, summarize_change, summarize_glm_count, summarize_num_patients, analyze_num_patients, summarize_patients_exposure_in_cols, coxph_pairwise, tabulate_survival_subgroups, surv_time, and surv_timepoint.

Enhancements

• Added formatting function format_count_fraction_lt10 for formatting count_fraction with special consideration when count is less than 10.
• Updated s_summary.logical output for count_fraction when denominator is zero to display as NA instead of 0 in tables.
• Updated analyze_vars_in_cols to allow character input to indicate whether nominal time point is post-dose or pre-dose when applying the 1/3 imputation rule.

Bug Fixes

• Fixed bug in g_km causing an error when converting certain annotation width units.

Miscellaneous

• Began deprecation of na_level argument in s_count_abnormal_by_baseline, a_summary, analyze_vars, analyze_vars_in_cols, compare_vars, h_map_for_count_abnormal, h_stack_by_baskets, summarize_colvars, a_coxreg, and summarize_coxreg and replaced it with the na_str argument.

What's Changed

• Add formatting function with special case when count less than 10 by @edelarua in #1075
• Expose na_str argument in analyze/summarize functions by @edelarua in #1071
• Fix s_summary return for empty logical vectors by @edelarua in #1079
• Fixes after refactor of string wrapping by @edelarua in #1080
• Fix g_km unit conversion bug by @edelarua in #1082
• 1086 update plot snaps@main by @ayogasekaram in #1089
• Update logo by @edelarua in #1091
• Fix logo sizing on pkgdown site by @edelarua in #1093
• Update analyze_vars _in_cols to allow non-numeric nominal timepoint variable by @edelarua in #1095
• update g_lineplot with cohort param. by @ayogasekaram in #1078
• Fix integration for version 9.1 by @Melkiades in #1097
• Revert failing g_km snapshot test by @edelarua in #1101
• remove USUBJID hard coding by @ayogasekaram in #1099
• Fix all lint by @edelarua in #1104
• [skip vbump] Release Candidate 0.9.2 by @Melkiades in #1107

Full Changelog: v0.9.1...v0.9.2