- Functions
estimate_KM()
andvisr.survfit()
have been deprecated in favor ofggsurvfit::ggsurvfit()
.
-
We now zoom in on a figure (e.g. Kaplan-Meier figure) with
ggplot2::coord_cartesian()
instead of usingscale_x_continuous(limits=)
andscale_y_continuous(limits=)
. The latter first removes data outside the limits, then constructs the line. Zooming constructs the full line, then zooms into the limits. This is useful because the risktable often reports estimates near the end of a KM figure, but the line is cutoff and not shown at the last time point. (#402) -
README update to contributor listing. (#435)
-
Highlight specific strata in survival plots using
add_highlight()
. -
Indicate quantiles in survival plots using
add_quantiles()
. -
Estimation of cumulative incidence in the presence of competing risks is now possible with
estimate_cuminc()
. The risk estimates can be plotted similarly to estimates fromestimate_KM()
using thevisr()
function. -
Function
Surv_CNSR()
used the CDISC ADTTE conventions for the default values of the time and status indicators for time-to-event analyses. Of note, the status variable must be coded as 0/1 with 1 indicating an observation has been censored. (#391)
AlignPlots()
has been renamed toalign_plots()
.
-
Fix in
get_pvalue()
forestimate_KM()
objects when data piped in with modifications (e.g.dplyr::filter()
,dplyr::mutate()
). -
Fix in
add_risktable()
with estimates being assigned the incorrect label. -
Review unit testing for
estimate_KM(formula=)
argument. (#399) -
Fixed bug in
define_theme(strata=)
. (#388) -
Fix a discrepancy issue when saving
survfit()
call inestimate_KM()
. (#365) -
Added check for no
class(x)==
. (#358) -
Fix the documentation of the S3 method
visr()
. (#301)
-
Stratifying variable names have been removed from legend in
visr.survfit()
figures, and the legend title now describes the stratifying variable(s). (#343) -
The
estimate_KM()
function gains an experimentalformula=
argument. When the argument is used, the AVAL, CNSR, and strata arguments typically used to construct the formula are ignored. (#379) -
add_risktable
has a rowgutter argument to allow spacing between plot and risktables -
The strata variable is now removed from the body of
tableone()
results. (#254) -
The
visr.survfit()
function no longer warns about x-axis label whenPARAM
column not found in original data set. (#378) -
The call saved in the
estimate_KM()
object has been updated to a quosure--ensuring the original function call can always be recalled. -
define_theme()
updated to ensure that only the strata present in the theme are displayed. (#388) -
Removed the external dependency with the
easyalluvial
package. (#383) -
Reduced the number of exported functions. (#381)
-
Improved documentation for
visr()
and other generic functions. (#301) (#357) -
Unit testing updated where relevant to have 100% code coverage and function requirements tested.
-
The
README
page has been updated with additional examples. (#425) -
lifecycle
badges have been added at the function level to indicate functions that are still inexperimental
orquestioning
phases. (#398)
- Initial CRAN release.