Skip to contents

Plot individual longitudinal trajectories (spaghetti plot)

Source: R/spaghettiPlot.R
spaghettiPlot.Rd

Draws one line per subject for a longitudinal biomarker, optionally colored by the subject's eventual outcome, with a smoothed mean trajectory per outcome group overlaid. With align = "event" the horizontal axis becomes the time remaining until the subject's event/censoring time (survival_variable - time_variable, so 0 is the event time, at the left edge). This mirrors the backward joint model's view of the biomarker as a function of time-to-event. Unlike cmtPlot, which averages over subjects sharing one event time, this shows every subject's raw history.

Usage

spaghettiPlot(
  data_plot_all,
  bio_variable,
  time_variable,
  id_variable = "id",
  survival_variable = NULL,
  event_type_variable = NULL,
  align = c("baseline", "event"),
  n_subjects = NULL,
  smooth = TRUE,
  alpha = 0.3
)

Arguments

data_plot_all

A long-format data.frame with one row per biomarker measurement.

bio_variable

Name of the biomarker variable to plot.

time_variable

Name of the measurement-time variable.

id_variable

Name of the subject ID column. Default is "id".

survival_variable

Name of the time-to-event variable. Required when align = "event"; otherwise ignored.

event_type_variable

Name of the variable used to color subjects (e.g. the event indicator or event type). Must be constant within each subject. NULL (default) draws every subject in one color.

align

Either "baseline" (default), plotting against time_variable, or "event", plotting against the time until the event, survival_variable - time_variable.

n_subjects

Optional number of subjects to draw, sampled at random (call set.seed() first for a reproducible sample). NULL (default) draws every subject. The smoothed means use the drawn subjects only.

smooth

Logical; overlay a LOESS mean trajectory per group. Default TRUE.

alpha

Transparency of the individual lines. Default 0.3.

Value

A ggplot object.

Examples

data(pbc3)

# trajectories colored by death (1) vs alive/transplanted (0)
spaghettiPlot(pbc3, bio_variable = "serBilir", time_variable = "year",
  event_type_variable = "status2")


# aligned at each subject's event/censoring time
spaghettiPlot(pbc3, bio_variable = "serBilir", time_variable = "year",
  survival_variable = "years", event_type_variable = "status2",
  align = "event")