Facet survival curves into multiple panels

[kassambara]

No description provided.

[kassambara]

New option facet.by added now in ggsurvplot() to draw multi-panel survival curves of a data set grouped by one or two variables. Alias of the ggsurvplot_facet() function.

Load required packages

library(survival)
library(survminer)

Compute survival curves

fit1 <- survfit( Surv(time, status) ~ sex, data = colon )

fit2 <- survfit( Surv(time, status) ~ sex + rx, data = colon )

Facet by one grouping variables

Facet by the treatment assignment rx

# Faceting fit 1
ggsurvplot(fit1, colon, facet.by = "rx", 
              palette = "jco", pval = TRUE)

# Faceting fit 2
ggsurvplot(fit2, colon, facet.by = "adhere", 
           palette = "jco", pval = TRUE)

Facet by two grouping variables

ggsurvplot(fit1, colon, facet.by = c("rx", "adhere"),
           palette = "jco", pval = TRUE)

[jiajiaybyb]

I realized ggsurvplot_facet does not plot the survival lines all the way to 100%, is there a solution for this? Ideally I would like to have facet panels, with corresponding p values and survival tables in the output. Any recommend method for this? Thanks a lot!

[Yatros]

Hello,

Is it possible to add capability to ggsurvplot_facet() so that it accepts as input not only survfit objects, but also a list of survfit objects?

ggsurvplot() accepts already survfit lists.

Thank you very much,

Best Regards,

[audyavar]

Thanks so much @kassambara for adding this functionality. I use it regularly. If possible, can you add functionality to loop thru multiple variables i.e. if I m interested in say 10 median-cut variables and generate ggsurvplots faceted by a clinical variable like age, sex.
for(i in cols){ # cols are 10 variables which are genes
fit <- surv_fit(os ~ get(i), data = su_df)
p <- ggsurvplot(fit, su_df, facet.by = "psa_bins",
palette = "jco", pval = TRUE)
print(p)
}
Error in grDevices::col2rgb(colour, TRUE) : invalid color name 'get(i)'
the facet option doesnt seem to like the get() function.