processSurvTerms {psichomics}R Documentation

Process survival curves terms to calculate survival curves

Description

Process survival curves terms to calculate survival curves

Usage

processSurvTerms(clinical, censoring, event, timeStart, timeStop = NULL,
  group = NULL, formulaStr = NULL, coxph = FALSE, scale = "days",
  followup = "days_to_last_followup", survTime = NULL)

Arguments

clinical

Data frame: clinical data

censoring

Character: censor using "left", "right", "interval" or "interval2"

event

Character: name of column containing time of the event of interest

timeStart

Character: name of column containing starting time of the interval or follow up time

timeStop

Character: name of column containing ending time of the interval (only relevant for interval censoring)

group

Character: group relative to each patient

formulaStr

Character: formula to use

coxph

Boolean: fit a Cox proportional hazards regression model? FALSE by default

scale

Character: rescale the survival time to "days", "weeks", "months" or "years"

followup

Character: name of column containing follow up time

survTime

survTime object: times to follow up, time start, time stop and event (optional)

Details

If survTime is NULL, the survival times will be fetch from the clinical dataset according to the names given in timeStart, timeStop, event and followup. This can became quite slow when using the function in a for loop. If these variables are constant, consider running the function getAttributesTime to retrieve the time of such columns once and hand the result to the survTime argument of this function.

Value

A list with a formula object and a data frame with terms needed to calculate survival curves

Examples

clinical <- read.table(text = "2549   NA ii  female
                                840   NA i   female
                                 NA 1204 iv    male
                                 NA  383 iv  female
                               1293   NA iii   male
                                 NA 1355 ii    male")
names(clinical) <- c("patient.days_to_last_followup", 
                     "patient.days_to_death",
                     "patient.stage_event.pathologic_stage",
                     "patient.gender")
timeStart  <- "days_to_death"
event      <- "days_to_death"
formulaStr <- "patient.stage_event.pathologic_stage + patient.gender"
survTerms  <- processSurvTerms(clinical, censoring="right", event, timeStart,
                               formulaStr=formulaStr)

[Package psichomics version 1.10.1 Index]