adjustedCurves (version 0.11.2)

FGR_MI: Fine & Gray Model with Multiple Imputation

Description

This function can be utilized to calculate Fine & Gray models for multiply imputed datasets.

Usage

FGR_MI(mids, formula, cause=1, ...)

Value

A mira object containing the FGR regression for every imputed dataset.

Arguments

mids

A mids object created using the mice function. This replaces the data argument in the original function call.

formula

A formula object passed to the FGR function in the riskRegression package.

cause

The failure type of interest. Defaults to 1.

...

Other arguments which should be passed to the FGR function in the riskRegression package.

Author

Robin Denz

Details

A small convenience function to calculate Fine & Gray models for multiply imputed data. It is simply a wrapper around the FGR function from the riskRegression package, because the usual use of with is not supported directly. It returns a mira object, which can be passed to the outcome_model argument inside of the adjustedcif function when needed. No pool method or other functionality is available.

See Also

adjustedsurv

Examples

# not run because it would be too slow
# \donttest{
library(adjustedCurves)
library(survival)

if (requireNamespace("riskRegression") & requireNamespace("prodlim") &
    requireNamespace("mice")) {

library(riskRegression)
library(mice)
library(prodlim)

# simulate some data as example
sim_dat <- sim_confounded_crisk(n=50, max_t=1.2)
sim_dat$group <- as.factor(sim_dat$group)

# introduce random missingness in x1 as example
sim_dat$x1 <- ifelse(runif(n=50) < 0.5, sim_dat$x1, NA)

# perform multiple imputation
mids <- mice::mice(data=sim_dat, method="pmm", m=5, printFlag=FALSE)

# use the function
fgr_mods <- FGR_MI(mids=mids,
                   formula=Hist(time, event) ~ x1 + x2 + x3 + x4 + x5 + x6 + group,
                   cause=1)
}
# }