KMunicate

Create KMunicate-Style Kaplan–Meier Plots

https://github.com/ellessenne/kmunicate-package

Repository

Create KMunicate-Style Kaplan–Meier Plots

README.Rmd

---
output: github_document
editor_options: 
  chunk_output_type: console
---



```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  fig.align = "center",
  out.width = "90%",
  dpi = 300,
  warning = FALSE,
  message = FALSE,
  dev = "ragg_png"
)
```

# KMunicate-Style Kaplan–Meier Plots


[![Codecov test coverage](https://codecov.io/gh/ellessenne/KMunicate-package/branch/master/graph/badge.svg)](https://app.codecov.io/gh/ellessenne/KMunicate-package?branch=master)
[![CRAN status](https://www.r-pkg.org/badges/version/KMunicate)](https://CRAN.R-project.org/package=KMunicate)
[![CRAN_Logs_Badge](http://cranlogs.r-pkg.org/badges/KMunicate)](https://cran.r-project.org/package=KMunicate)
[![CRAN_Logs_Badge_Total](http://cranlogs.r-pkg.org/badges/grand-total/KMunicate)](https://cran.r-project.org/package=KMunicate)
[![R-CMD-check](https://github.com/ellessenne/KMunicate-package/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/ellessenne/KMunicate-package/actions/workflows/R-CMD-check.yaml)


The goal of {KMunicate} is to produce Kaplan–Meier plots in the style recommended following the [KMunicate study](http://dx.doi.org/10.1136/bmjopen-2019-030215) (TP Morris *et al*. Proposals on Kaplan–Meier plots in medical research and a survey of stakeholder views: KMunicate. *BMJ Open*, 2019, 9:e030215). 

## Installation

You can install {KMunicate} from [CRAN](https://CRAN.R-project.org/package=KMunicate) by typing the following in your R console:

``` r
install.packages("KMunicate")
```

Alternatively, you can install the dev version of {KMunicate} from [GitHub](https://github.com/ellessenne/KMunicate-package/) with:

``` r
# install.packages("devtools")
devtools::install_github("ellessenne/KMunicate-package")
```

## Example

```{r packages}
library(survival)
library(KMunicate)
```

The {KMunicate} package comes with a couple of bundled dataset, `cancer` and `brcancer`.
The main function is named `KMunicate`:

```{r brcancer, fig.height = 7, fig.width = 7 / sqrt(2)}
KM <- survfit(Surv(rectime, censrec) ~ hormon, data = brcancer)
time_scale <- seq(0, max(brcancer$rectime), by = 365)
KMunicate(fit = KM, time_scale = time_scale)
```

```{r cancer, fig.height = 7, fig.width = 7 / sqrt(2)}
KM <- survfit(Surv(studytime, died) ~ drug, data = cancer2)
time_scale <- seq(0, max(cancer2$studytime), by = 7)
KMunicate(fit = KM, time_scale = time_scale)
```

You also might wonder, does this work with a single arm?
Yes, yes it does:

```{r cancer-single, fig.height = 6, fig.width = 6}
KM <- survfit(Surv(studytime, died) ~ 1, data = cancer2)
time_scale <- seq(0, max(cancer2$studytime), by = 7)
KMunicate(fit = KM, time_scale = time_scale)
```

Finally, you can also plot 1 - survival by using the argument `.reverse = TRUE`:

```{r brcancer-reverse, fig.height = 7, fig.width = 7 / sqrt(2)}
KM <- survfit(Surv(rectime, censrec) ~ hormon, data = brcancer)
time_scale <- seq(0, max(brcancer$rectime), by = 365)
KMunicate(fit = KM, time_scale = time_scale, .reverse = TRUE)
```

## Customise Risk Table

By default, `KMunicate()` will build a risk table conform to the KMunicate style, e.g., with cumulative number of events and censored (the column-wise sum is equal to the total number of individuals at risk per arm):

```{r brcancer-KMunicate, fig.height = 7, fig.width = 7 / sqrt(2)}
KM <- survfit(Surv(rectime, censrec) ~ hormon, data = brcancer)
time_scale <- seq(0, max(brcancer$rectime), by = 365)
KMunicate(fit = KM, time_scale = time_scale)
```

Alternatively, it is possible to customise the risk table via the `.risk_table` argument. 
For instance, if one wants to have interval-wise number of events and censored, just pass the `survfit` value to the `.risk_table` argument:

```{r brcancer-survfit, fig.height = 7, fig.width = 7 / sqrt(2)}
KMunicate(fit = KM, time_scale = time_scale, .risk_table = "survfit")
```

This is the default output of the `summary.survfit()` function.

Finally, it is also possible to fully omit the risk table by setting `.risk_table = NULL`:

```{r brcancer-NULL, fig.height = 6 / sqrt(2), fig.width = 6}
KMunicate(fit = KM, time_scale = time_scale, .risk_table = NULL)
```

## Custom Fonts

Assuming you have set up your computer to use custom fonts with `ggplot2`, customising your KMunicate-style plot is trivial.
All you have to do is pass the font name as the `.ff` argument:

```{r cancer-single-ff, fig.height = 6, fig.width = 6}
KM <- survfit(Surv(studytime, died) ~ 1, data = cancer2)
time_scale <- seq(0, max(cancer2$studytime), by = 7)
KMunicate(fit = KM, time_scale = time_scale, .ff = "Times New Roman")
```

## Further Customisation

Several options to further customise each plot are provided, see e.g. the introductory vignette for more details.

Owner

GitHub Events

Committers

Last synced: over 2 years ago

Issues and Pull Requests

Last synced: over 2 years ago

Packages