pred2.crr: Predict Cumulative Incidence Rate
In jixccf/QHScrnomo: Construct nomograms for competing risk models

Description Usage Arguments Details Value Author(s) See Also Examples

Predict Cumulative Incidence Rate

1	pred2.crr(f.crr, lp, time)

`f.crr`	a saved competing risks regression model created by function `crr.fit`
`lp`	a scalar being the sum of linear predictors for a single subject.
`time`	expected time point, at which cumulative incidence rate will be assessed.

Calculate the predicted cumulative incidence rate based on a saved competing risks regression model. The cumulative incidence is adjusted for other competing causes rather than the event of interest.

Return the predicted cumulative incidence rate.

Michael Kattan, Ph.D, Changhong Yu
Department of Quantitative Health Sciences
Cleveland Clinic

predict.crr crr.fit

data(prostate.dat)
library(Hmisc,TRUE)
library(rms,TRUE)
dd <- datadist(prostate.dat)
options( datadist = "dd")
f.cph <- cph(formula = Surv(TIME_EVENT, EVENT_DOD == 1 ) ~ rcs(AGE,3) +
    CLIN_STG + rcs(PSA, 3),
    data = prostate.dat, x = TRUE, y = TRUE, surv = TRUE)
f.crr <- crr.fit(fit = f.cph, cencode = 0, failcode = 1)

# Estimate cumulative incidence rate by 6 year
QHScrnomo:::pred2.crr(prostate.crr, lp = 0.8, time = 60)