Title: Kaplan-Meier Plot with 'ggplot2'
Version: 0.5.13
Date: 2025-06-11
Description: The function 'jskm()' creates publication quality Kaplan-Meier plot with at risk tables below. 'svyjskm()' provides plot for weighted Kaplan-Meier estimator.
Depends: R (≥ 3.4.0)
License: Apache License 2.0
Encoding: UTF-8
Imports: ggplot2, ggpubr, survival, survey, scales, patchwork, cmprsk, stats
RoxygenNote: 7.3.2
URL: https://github.com/jinseob2kim/jskm, https://jinseob2kim.github.io/jskm/
BugReports: https://github.com/jinseob2kim/jstable/issues
Suggests: testthat, knitr, rmarkdown
VignetteBuilder: knitr
NeedsCompilation: no
Packaged: 2025-06-12 13:38:20 UTC; js
Author: Jinseob Kim ORCID iD [aut, cre], yoonkyoung Chun [aut], Zarathu [cph, fnd], sungho Choi [aut]
Maintainer: Jinseob Kim <jinseob2kim@gmail.com>
Repository: CRAN
Date/Publication: 2025-06-12 15:20:06 UTC

Adjusted Log-Rank Test

Description

Performs an adjusted log-rank test considering weights.

Usage

adjusted.LR(times, failures, variable, weights = NULL)

Arguments

times

A numeric vector of survival times.

failures

A binary vector indicating event occurrence.

variable

A numeric binary variable (0 and 1) indicating group membership.

weights

A numeric vector of weights.

Value

A list containing the test statistic and p-value.

Creates a Kaplan-Meier plot for survfit object.

Description

Creates a Kaplan-Meier plot with at risk tables below for survfit object.

Usage

jskm(
  sfit,
  table = FALSE,
  table.censor = FALSE,
  xlabs = "Time-to-event",
  ylabs = NULL,
  xlims = c(0, max(sfit$time)),
  ylims = c(0, 1),
  surv.scale = c("default", "percent"),
  ystratalabs = NULL,
  ystrataname = "Strata",
  timeby = signif(max(sfit$time)/7, 1),
  main = "",
  pval = FALSE,
  pval.size = 5,
  pval.coord = c(NULL, NULL),
  pval.testname = T,
  marks = TRUE,
  shape = 3,
  med = FALSE,
  legend = TRUE,
  legendposition = c(0.85, 0.8),
  ci = FALSE,
  subs = NULL,
  label.nrisk = "Numbers at risk",
  size.label.nrisk = 10,
  linecols = "Set1",
  dashed = FALSE,
  cumhaz = F,
  cluster.option = "None",
  cluster.var = NULL,
  data = NULL,
  cut.landmark = NULL,
  showpercent = F,
  status.cmprsk = NULL,
  linewidth = 0.75,
  theme = NULL,
  nejm.infigure.ratiow = 0.6,
  nejm.infigure.ratioh = 0.5,
  nejm.infigure.ylim = c(0, 1),
  surv.by = NULL,
  nejm.surv.by = NULL,
  hr = FALSE,
  hr.size = 5,
  hr.coord = c(NULL, NULL),
  hr.testname = F,
  ...
)

Arguments

sfit

a survfit object

table

logical: Create a table graphic below the K-M plot, indicating at-risk numbers?

table.censor

logical: Add numbers of censored in table graphic

xlabs

x-axis label

ylabs

y-axis label

xlims

numeric: list of min and max for x-axis. Default = c(0,max(sfit$time))

ylims

numeric: list of min and max for y-axis. Default = c(0,1)

surv.scale

scale transformation of survival curves. Allowed values are "default" or "percent".

ystratalabs

character list. A list of names for each strata. Default = names(sfit$strata)

ystrataname

The legend name. Default = "Strata"

timeby

numeric: control the granularity along the time-axis; defaults to 7 time-points. Default = signif(max(sfit$time)/7, 1)

main

plot title

pval

logical: add the pvalue to the plot?

pval.size

numeric value specifying the p-value text size. Default is 5.

pval.coord

numeric vector, of length 2, specifying the x and y coordinates of the p-value. Default values are NULL

pval.testname

logical: add '(Log-rank)' text to p-value. Default = F

marks

logical: should censoring marks be added?

shape

what shape should the censoring marks be, default is a vertical line

med

should a median line be added to the plot? Default = F

legend

logical. should a legend be added to the plot?

legendposition

numeric. x, y position of the legend if plotted. Default=c(0.85,0.8)

ci

logical. Should confidence intervals be plotted. Default = FALSE

subs

= NULL,

label.nrisk

Numbers at risk label. Default = "Numbers at risk"

size.label.nrisk

Font size of label.nrisk. Default = 10

linecols

Character or Character vector. Colour brewer pallettes too colour lines. Default ="Set1", "black" for black with dashed line, character vector for the customization of line colors.

dashed

logical. Should a variety of linetypes be used to identify lines. Default = FALSE

cumhaz

Show cumulative incidence function, Default: F

cluster.option

Cluster option for p value, Option: "None", "cluster", "frailty", Default: "None"

cluster.var

Cluster variable

data

select specific data - for reactive input, Default = NULL

cut.landmark

cut-off for landmark analysis, Default = NULL

showpercent

Shows the percentages on the right side.

status.cmprsk

Status value when competing risk analysis, Default = 2nd level of status variable

linewidth

Line witdh, Default = 0.75

theme

Theme of the plot, Default = NULL, "nejm" for NEJMOA style, "jama" for JAMA style

nejm.infigure.ratiow

Ratio of infigure width to total width, Default = 0.6

nejm.infigure.ratioh

Ratio of infigure height to total height, Default = 0.5

nejm.infigure.ylim

y-axis limit of infigure, Default = c(0,1)

surv.by

breaks unit in y-axis, default = NULL(ggplot default)

nejm.surv.by

breaks unit in y-axis in nejm figure, default = NULL(ggplot default)

hr

logical: add the hazard ratio to the plot?

hr.size

numeric value specifying the HR text size. Default is 5.

hr.coord

numeric vector, of length 2, specifying the x and y coordinates of the p-value. Default values are NULL

hr.testname

logical: add '(Log-rank)' text to p-value. Default = F

...

PARAM_DESCRIPTION

Details

DETAILS

Value

Plot

Author(s)

Jinseob Kim, but heavily modified version of a script created by Michael Way. https://github.com/michaelway/ggkm/ I have packaged this function, added functions to namespace and included a range of new parameters.

Examples

library(survival)
data(colon)
fit <- survfit(Surv(time, status) ~ rx, data = colon)
jskm(fit, timeby = 500)

Creates a Weighted Kaplan-Meier plot - svykm.object in survey package

Description

Creates a Weighted Kaplan-Meier plot - svykm.object in survey package

Usage

svyjskm(
  sfit,
  theme = NULL,
  xlabs = "Time-to-event",
  ylabs = "Survival probability",
  xlims = NULL,
  ylims = c(0, 1),
  ystratalabs = NULL,
  ystrataname = NULL,
  surv.scale = c("default", "percent"),
  timeby = NULL,
  main = "",
  pval = FALSE,
  pval.size = 4,
  pval.coord = c(NULL, NULL),
  pval.testname = F,
  marks = FALSE,
  hr = FALSE,
  hr.size = 2,
  hr.coord = c(NULL, NULL),
  med = FALSE,
  legend = TRUE,
  legendposition = c(0.85, 0.8),
  ci = NULL,
  linecols = "Set1",
  dashed = FALSE,
  cumhaz = F,
  design = NULL,
  subs = NULL,
  table = F,
  table.censor = F,
  label.nrisk = "Numbers at risk",
  size.label.nrisk = 10,
  cut.landmark = NULL,
  showpercent = F,
  linewidth = 0.75,
  nejm.infigure.ratiow = 0.6,
  nejm.infigure.ratioh = 0.5,
  nejm.infigure.ylim = c(0, 1),
  surv.by = NULL,
  nejm.surv.by = NULL,
  ...
)

Arguments

sfit

a svykm object

theme

Theme of the plot, Default = NULL, "nejm" for NEJMOA style, "jama" for JAMA style

xlabs

x-axis label, Default: 'Time-to-event'

ylabs

y-axis label.

xlims

numeric: list of min and max for x-axis. Default: NULL

ylims

numeric: list of min and max for y-axis. Default: c(0, 1)

ystratalabs

character list. A list of names for each strata. Default: NULL

ystrataname

The legend name. Default: 'Strata'

surv.scale

scale transformation of survival curves. Allowed values are "default" or "percent".

timeby

numeric: control the granularity along the time-axis; defaults to 7 time-points.

main

plot title, Default: ”

pval

logical: add the pvalue to the plot?, Default: FALSE

pval.size

numeric value specifying the p-value text size. Default is 4.

pval.coord

numeric vector, of length 2, specifying the x and y coordinates of the p-value. Default values are NULL

pval.testname

logical: add '(Log-rank)' text to p-value. Default = F

marks

logical: should censoring marks be added?

hr

logical: add the Hazard Ratio to the plot?, Default: FALSE

hr.size

numeric value specifying the Hazard Ratio text size. Default is 2.

hr.coord

numeric vector, of length 2, specifying the x and y coordinates of the Hazard Ratio. Default values are NULL

med

should a median line be added to the plot? Default = F

legend

logical. should a legend be added to the plot?

legendposition

numeric. x, y position of the legend if plotted. Default=c(0.85,0.8)

ci

logical. Should confidence intervals be plotted. Default = NULL

linecols

Character or Character vector. Colour brewer pallettes too colour lines. Default ="Set1", "black" for black with dashed line, character vector for the customization of line colors.

dashed

logical. Should a variety of linetypes be used to identify lines. Default: FALSE

cumhaz

Show cumulaive incidence function, Default: F

design

Data design for reactive design data , Default: NULL

subs

= NULL,

table

logical: Create a table graphic below the K-M plot, indicating at-risk numbers?

table.censor

logical: Add numbers of censored in table graphic

label.nrisk

Numbers at risk label. Default = "Numbers at risk"

size.label.nrisk

Font size of label.nrisk. Default = 10

cut.landmark

cut-off for landmark analysis, Default = NULL

showpercent

Shows the percentages on the right side.

linewidth

Line witdh, Default = 0.75

nejm.infigure.ratiow

Ratio of infigure width to total width, Default = 0.6

nejm.infigure.ratioh

Ratio of infigure height to total height, Default = 0.5

nejm.infigure.ylim

y-axis limit of infigure, Default = c(0,1)

surv.by

breaks unit in y-axis, default = NULL(ggplot default)

nejm.surv.by

breaks unit in y-axis in nejm figure, default = NULL(ggplot default)

...

PARAM_DESCRIPTION

Details

DETAILS

Value

plot

Examples

library(survey)
data(pbc, package = "survival")
pbc$randomized <- with(pbc, !is.na(trt) & trt > 0)
biasmodel <- glm(randomized ~ age * edema, data = pbc)
pbc$randprob <- fitted(biasmodel)
dpbc <- svydesign(id = ~1, prob = ~randprob, strata = ~edema, data = subset(pbc, randomized))
s1 <- svykm(Surv(time, status > 0) ~ sex, design = dpbc)
svyjskm(s1)