Skip to contents

Forest plot of the filed analysis and the supplementary rows

Source: R/display.R
plot_forest.Rd

Estimate and 95% interval for the filed analysis and the named rows of each census, one panel per endpoint-metric row, the filed row first. When several results objects are given (a trial alone and pooled with an earlier trial, say), the objects are drawn side by side within each row in different colours.

Usage

plot_forest(..., rows = NULL, xlab = "Treatment difference (95% interval)")

Arguments

...

One or more stress_census or stress_result objects (or lists of them, as in plot_swarm()). Several objects are compared side by side; name the arguments to label them.

rows

Character vector of census row labels to show. The default is every counted row with role = "supplementary".

xlab

Axis label.

Value

A ggplot object.

Examples

spec <- stress_spec(mmrm::fev_data, "FEV1", "USUBJID", "AVISIT", "ARMCD", "PBO",
                    covariates = ~ RACE + SEX + FEV1_BL)
d <- mmrm::fev_data
cen <- stress_census(spec, stress_covariance(spec, "cs"),
                     stress_data(spec, list("Black and White only" = d[d$RACE != "Asian", ],
                                            "Female only" = d[d$SEX == "Female", ])))
if (requireNamespace("ggplot2", quietly = TRUE)) {
  plot_forest(cen)
  plot_forest("All" = cen, "Complete cases" = cen, rows = "Black and White only")
}