gsDesign2 1.1.8

New features

• The gs_design_ahr() function has been updated to allow outputting spending time (#602, thanks to @LittleBeannie).

Documentation

• Documentation has been migrated to roxygen2 7.3.3 for improved maintainability and consistency (#601, thanks to @jdblischak).
• An example demonstrating how to update a stratified design at the time of analysis has been added to the gs_update_ahr() function documentation (#608, thanks to @LittleBeannie).

Testing

• Additional developer tests of gs_design_ahr() have been added (#604, thanks to @LittleBeannie).

gsDesign2 1.1.7

Documentation

• The vignette discussing the futility boundary design is finalized (#583, thanks to @LittleBeannie)
• An example calculating expected events for multiple strata is added (#597, thanks to @LittleBeannie)
• The documentation of common parameters is unified by @inheritparams (#598, thanks to @LittleBeannie)
• A contributing guide with overview of package architecture is added (#596, thanks to @jdblischak)

Programming improvements

• The S3 classes are re-factored (#584, thanks to @jdblischak)
• The input parameters are not displayed when printing the objects (#586, thanks to @jdblischak)
• text_summary() supports design objects with spending functions specified as character strings (#587, thanks to @yihui).
• The attribute assignment uninteger_is_from is fixed in to_integer() (#591, thanks to @jdblischak)
• text_summary() is updated to support fixed designs from gs_design_ahr() (#592, thanks to @jdblischak)

gsDesign2 1.1.6

Statistical improvements

• The gs_power_wlr() function now includes an h1_spending argument, allowing users to specify a spending under the alternative hypothesis (#565, thanks to @LittleBeannie).
• The following functions now support info_scale argument: fixed_design_ahr(), fixed_design_fh(), fixed_design_mb(), and fixed_design_rd() (#571, thanks to @LittleBeannie).
• Functions for fixed designs with integer sample sizes now return the average HR in their output, providing a more complete summary of the design characteristics. (#572, thanks to @LittleBeannie).

Documentation

• The documentation for gs_design_npe() and gs_power_npe() has been consolidated into a single topic for improved clarity and easier navigation. (#567, thanks to @LittleBeannie).
• The package codebase has been updated to use the native R pipe (|>) exclusively, removing the magrittr dependency and aligning with modern R practices. (#577, thanks to @jdblischak).

gsDesign2 1.1.5

Bug fixes

• The spending of the WLR design has been corrected. The default spending of a WLR design is information fraction. (#557, thanks to @LittleBeannie).

gsDesign2 1.1.4

New functions

• A new function, gs_cp_npe(), is now available for calculating simple conditional power under NPH. A vignette has been published on the pkgdown website (#510, #539, #545, thanks to @LittleBeannie).
• A new design summary function gs_bound_summary() is available with similar output structure as gsDesign::gsBoundSummary() and supporting for multiple alpha (#468, #522, #537, thanks to @jdblischak).
• Textual summary of AHR designs are available via the text_summary() (#526, thanks to @LittleBeannie).

User interface improvements

• The integer event is rounded in IAs and rounded up in FA as defaults in to_integer() (#483, thanks to @LittleBeannie).
• The integer sample size is rounded to the nearest multiple of randomization ratio + 1 when round_up_final = TRUE (#488, thanks to @LittleBeannie)..
• The updated design can be derived with the input of events per analysis per interval (#499, thanks to @LittleBeannie).
• Bound functions and spending functions can be provided through their names (character strings) now, e.g., gs_design_ahr(..., upper = "gs_spending_bound", upar = list(sf = "sfLDOF", ...)) (#509, thanks to @yihui).
• The footnote argument of as_gt() can take the value FALSE to disable footnotes (#514, thanks to @yihui).
• Optimized the functions expected_accural(), expected_time(), gs_design_ahr(), gs_design_combo(), gs_design_npe(), gs_design_wlr(), pw_info(), ppwe(), s2pe(), and gs_bound() (#528, thanks to @yihui).
• Weights of a WLR design functions can be provided as string now (#533, thanks to @LittleBeannie).

Bug fixes

• Correct the statistical information of WLR-integer designs in to_integer() (#478, thanks to @LittleBeannie).
• Correct the calculation info_frac in gs_design_wlr() when info_scale = "h0_info" (#485, #486, thanks to @LittleBeannie).
• Add the h1_spending argument to gs_power_ahr() (#518, thanks to @LittleBeannie)

Testing

• Add the developer tests of gs_cp_npe() (#519, thanks to @shiyuskaya)

gsDesign2 1.1.3

Bug fixes

• Fix the bug of gs_design_rd() when info_scale = "h0_h1_info" (#402, thanks to @LittleBeannie).
• Fix the bug of gs_spending_combo() to enable HSD spending function (#444, thanks to @LittleBeannie).
• Fix the bug of fixed_design_maxcombo() regarding the upper bounds (#445, thanks to @elong0527).
• Fix the bug of gs_design_wlr() when the design is driven by information fraction only (#446, thanks to @LittleBeannie).
• Fix the bug of pw_info() when there are many piecewise HRs (#460, thanks to @LittleBeannie).

Statistical improvements

• The gs_update_ahr() function (test version) is updated to

  • Enable unchanged futility bounds when it is fixed at the original design (#408, thanks to @LittleBeannie).
  • Allow boundary updates when only certain interim analysis data is available (#436, thanks to @LittleBeannie).
  • Allow boundary updates when alpha changes by keeping the same info_scale as the original design (#470, @LittleBeannie).

• Rounding of integer design is updated (#488, #484, #486, thanks to @LittleBeannie).

• Integer design (i.e., integer sample size and events) is updated to ensure exact integer sample size and events (#452, #460, thanks to @LittleBeannie and @yihui).

• Change the information fraction displayed at the summary-gt table from under H1 to H0 for logrank tests (#439, thanks to @LittleBeannie).

• Add the sample size as the output of ahr() and pw_info() (#427, #433, thanks to @LittleBeannie).

User interface improvements

• Enable passing named vector of col_decimals and analysis_decimals to summary.gs_design() (#403, #431, @jdblischak).

Coding practice improvements

• Add robust check if lower is equivalent to gs_b (#413, thanks to @jdblischak )
• The summary(), as_gt() as_rtf(), and to_integer() functions are refactored (#448, #449, #450, #465, #461, thanks to @yihui).
• Remove the full_alpha argument from as_rtf.gs_design() (#458, thanks to @yihui).
• Other coding practice improvement worth mentioning (#409, #412, #429, #437, #440, #447, #453, #464, #467, $475, thanks to @yihui, @nanxstats, and @jdblischak).

Documentation

• Update the documentation of gs_b() (#415, @jdblischak)

Testing

• More developer tests of gs_power_ahr() are added (#420, @LittleBeannie).
• More developer tests of summary() are added (#422, #426, thanks to @yuliasidi, @jdblischak and @LittleBeannie).
• Independent tests of ahr_blinded() are added (#435, thanks to @DMuriuki).
• More developer tests of to_integer() are added (#476, thanks to @LittleBeannie).

gsDesign2 1.1.2

New features

• The gs_update_ahr() function is now available for efficacy and futility boundary update based on blinded estimation of treatment effect (#370).

Bug fixes

• Fix the accrual parameters bugs in gs_design_wlr() depending on npsurvSS (#344, #356).
• Fix gs_design_ahr() to incorporate information fraction driven design when number of analyses >= 4 (#358).

Improvements

• Zero failure rate in some but not all intervals is acceptable as input (#360).
• Study with duration > 100 units are executable when event accrual is slow (#368).

Documentation

• A new vignette introducing how to do the boundary update is available (#278, #364, #366).
• A new vignette bridging gsDesign2 to the 6 test types of gsDesign is available.
• The pkgdown website is re-organized to providing better view for users (#341).

Testing

• Independent testing of as_gt() is added (#337).
• Restructure tests to make them self-contained (#347).

gsDesign2 1.1.1

New features

• The as_rtf() method is now available for fixed_design and gs_design objects for generating RTF table outputs (#278).

Bug fixes

• gs_power_wlr() and to_integer() now check and convert integer sample size more rigorously (#322).
• gs_design_*() now handle exceptions explicitly when all hazard ratio is set to 1 throughout the study (#301).
• fixed_design_rd() will not generate warnings due to the previous default value change of h1_spending (#296).

Improvements

• gs_power_ahr() now runs twice as fast by using data.table and other performance optimizations (#295), enhanced by similar improvements in gs_info_ahr() and pw_info() (#300).
• Enrollment and failure rate input constructors and validators are refactored to check only the format instead of the class. This change reduces the number of warning messages and catches real exceptions as errors properly (#316).
• Nested functions are refactored into reusable internal functions, to improve code rigor, avoid potential scoping pitfalls, and facilitate debugging (#235).
• For fixed designs, the variable names of the table outputs from to_integer() and summary() are updated (#292).

Documentation

• Add a new vignette statistical information under null and alternative hypothesis (#289).
• Improve define_enroll_rate() and define_fail_rate() documentation by adding detailed descriptions and improving code examples (#302).
• The function reference page now has dedicated sections for piecewise exponential distributions and computing trial events (#258).
• Use the four trailing dashes convention to standardize code comment section format (#308).

Namespace and testing

• Tidy up namespace by removing rlang from and adding stats to Imports (#307, #325).
• Qualify namespaces in tests to avoid library() calls (#332).
• Fortify the GitHub Actions workflows by limiting the token usage only when necessary and enabling manual trigger of workflow runs (#326).
• Update GitHub Actions workflows to the latest versions from upstream (#330).

gsDesign2 1.1.0

Breaking changes

• Split fixed_design() into a group of fixed_design_*() functions for enhanced modularity (#263).
• gs_design_rd() and gs_power_rd() now have updated options of weighting for stratified design (#276).
• ppwe() now accepts two arguments duration and rate instead of a data frame fail_rate (#254).
• Unexport helper functions gridpts(), h1(), and hupdate() (#253).

New features

• Introduce define_enroll_rate() and define_fail_rate() as new input constructor functions to replace the tibble inputs (#238).
• Add a new function pw_info() which calculates the statistical information under the piecewise model (#262).

Improvements

• Add a vignette showing the canonical joint distribution of Z-score and B-values under null and alternative hypothesis for the AHR test (#246).
• Refactor expected_event() to improve computational performance (@jdblischak, #250).
• Move the source code of the legacy version from inst/ to tests/testthat/ as developer tests (#269).

gsDesign2 1.0.9

Improvements

• Add CRAN download counts badge (#215).
• Update documentation of gs_design_rd() (#220).
• Format footnote numbers using decimal notation (#222).
• Split C++ functions into individual .cpp and header files (#224).

Bug fixes

• Fix the digits display in summary() (#231).

gsDesign2 1.0.8

Improvements

• Update the calculation of upper/lower bounds at the final analysis in MaxCombo tests (#217).
• Update the fixed_design() function in the application of stratified design when using the Lachin and Foulkes method (#211).
• Correct the fixed_design() function in the application of rmst (#212).
• Rename the info_scale argument options from c(0, 1, 2) to c("h0_h1_info", "h0_info", "h1_info") to be more informative and make the default value ("h0_h1_info") clear (#203).
• Add missing global functions/variables (#213).
• Fix outdated argument names and use canonical style for text elements in README.md (#198).
• Add a CRAN downloads badge to README.md to show the monthly downloads (#216).

Bug fixes

• Fix the calculation of the futility bounds in gs_power_ahr() (#202).

gsDesign2 1.0.7

Improvements

• Move imported dependencies from Suggests to Imports.
• Remove redundant dependencies from Suggests.
• Update the GitHub Actions workflows to their latest versions from upstream.
• Add a rule to .gitattributes for GitHub Linguist to keep the repository’s language statistics accurate.

gsDesign2 1.0.6

Improvements

• Export functions gridpts(), h1(), hupdate(), and gs_create_arm() to avoid the use of ::: in code examples.
• Fix the write path issue by moving the test fixture generation script to data-raw/ which is not included in the package.

gsDesign2 1.0.5

First submission to CRAN in March 2023.

Breaking changes

• Passes lintr check for the entire package (#150, #151, #171).
• Improve the documentation (#161, #163, #168, #176).

Bug fixes

• check_fail_rate() when only 1 number in fail_rate is > 0 (#132).
• gs_power_ahr() when study duration is > 48 months (#141).
• fixed_design() for event-based design (#143).
• gs_design_combo() when test only applies to part of the analysis (#148).
• gs_info_rd() for variance calculation (#153).
• summary() for capitalized first letter in the summary header (#164).

gsDesign2 1.0.0

GitHub release in December 2022.

Breaking changes

• Merges gsDesign2 v0.2.1 and gsdmvn.
• Updates API to follow the new style guide in vignette("style"). See the detailed mapping between the old API and new API in #84.

New features

• Supports organized summary tables and gt tables.
• Power/sample size calculation for risk difference.
• Integer sample size support (#116, #125).
• Adds fixed_design() to implement different methods for power/sample size calculation.
• Adds info_scale arguments to gs_design_*() and gs_power_*().
• Adds RMST and milestone methods to fixed design.

Bug fixes

• expected_accrual() for stratified population.
• gs_spending_bound() when IA is close to FA (#40).
• gs_power_bound() when applied in the MaxCombo test (#62).
• gs_design_npe() for type I error (#59).

Minor improvements

• Adds and re-organizes vignettes.

gsDesign2 0.2.1

GitHub release in August 2022.

• The release before merging with Merck/gsdmvn.

gsDesign2 0.2.0

GitHub release in May 2022.

• Supports the Biometrical Journal paper “A unified framework for weighted parametric group sequential design” by Keaven M. Anderson, Zifang Guo, Jing Zhao, and Linda Z. Sun.

gsDesign2 0.1.0

GitHub release in May 2021.

• Updated AHR vignette to introduce average hazard ratio concept properly.
• Added arbitrary distribution vignette to demonstrate s2pwe().
• Corrected calculations in AHR() when using stratified population.
• Release for Regulatory/Industry Symposium training.

gsDesign2 0.0.0.9006

GitHub release in December 2019.

• Added vignette for eEvents_df() explaining the methods thoroughly.
• Updated eEvents_df() to simplify output under option simple = FALSE.

gsDesign2 0.0.0.9005

GitHub release in December 2019.

• Updated docs/ directory to correct the reference materials on the website.
• Minor fixes in eAccrual().

gsDesign2 0.0.0.9004

GitHub release in November 2019.

• Moved new simulation functions to the simtrial package (simfix(), simfix2simPWSurv(), pMaxCombo()).

gsDesign2 0.0.0.9003

GitHub release in November 2019.

• Tried to make AHR() and simfix() more compatible with each other.
• Improved vignette for group sequential design.
• Added pkgdown website for documentation and vignettes.
• Added support functions for to support approximation using and visualization of the piecewise model.

gsDesign2 0.0.0.2

GitHub release in October 2019.

• Update AHR() to output trial duration, expected events and average hazard ratio in a tibble.
• Vignette AHRvignette demonstrating sample size computations for fixed design under non-proportional hazards assumptions.
• Vignette gsNPH demonstrating sample size computations for group sequential design under non-proportional hazards assumptions.
• Initial implementation of pMaxCombo() to compute p-value for MaxCombo test; pMaxComboVignette demonstrates this capability.

gsDesign2 0.0.0.1

GitHub release in September 2019.

• Computations based on piecewise constant enrollment and piecewise exponential failure rate.
• Expected event count calculation for each different hazard ratios in eEvents_df().
• Average hazard ratio computation based on expected event counts in AHR().
• Vignette demonstrating fixed sample size computation with simulation to verify power.