forest_plot: Forest Plot

In sjacobson94/clinicaltrials: A Collection of Functions for use in CCS

Description

This function creates the data for and plots a forest plot comparing hazard rates for categorical variables by a 2 level by variable.

Usage

forest_plot(
  data,
  stat,
  time,
  byvar,
  covars,
  ref = as.character(levels(data2$arm)[1]),
  labels = NULL,
  graph.pos = 3,
  xaxis.breaks = 0.25,
  title = "Hazard Rates",
  x.lab = "Hazard Ratios",
  plot.include = TRUE,
  x.max = max(values$upper, na.rm = TRUE),
  ...
)

Arguments

data	A dataset containing the necessary survival variables (event, time, other covariables).
stat	The event survival variable, numeric, with 1 coded as an event and 0 a censor.
time	The time survival variable, numeric.
byvar	The by variable with 2 distinct levels.
covars	A vector of categorical covariates that are desired for analysis.
ref	Reference by variable level (optional). Use this if specific by variable level is desired as the reference.
labels	An optional vector, soon to be list, of labels for the covariates in the 'covars' variable. Currently must be the same length as the 'covars' vector.
graph.pos	Column position for the graph.
xaxis.breaks	Distance between the x-axis breaks of the forest plot.
title	Title for the plot.
x.lab	Label for x-axis of plot.
plot.include	Logical to have plot return from function call.
x.max	Max value for x-axis. CI bands for intervals that pass this value will have an arrow on the right indicating a continuing band.
...	Additional arguments for the 'forestplot' function.

Value

This function returns a list of the forestplot produced and tables necessary to reproduce that plot.

Examples

library(arsenal)
data(mockstudy)
mockstudy %>%
 mutate(fu.stat = fu.stat - 1) %>%
 filter(arm != "A: IFL" & age.ord != "10-19" & race != "Other") %>%
 forest_plot(data = ., stat = "fu.stat", time = "fu.time", byvar = "arm", 
             covars = c("sex", "age.ord"), title = "Overall Survival", 
             labels = c("Gender", "Age Groups"), 
             xaxis.breaks = .5)

sjacobson94/clinicaltrials documentation built on Oct. 27, 2020, 6:43 p.m.