listLOS: Length of stay multistate model statistics

In n8thangreen/HESmanip: Hospital length of stay data preprocessing and analysis

Description

Calculate length of stay multistate model statistics.

Usage

listLOS(survData, tra, type, standerr = FALSE, trim = 0.99)

Arguments

survData
tra	Transition matrix of permitted direct between state jumps.
type	Model type, either cause-specific, subsdistribution with discharge alive as event of interest or in-hospital death as event of interest.
standerr	Logical, whether the standard error should be calculated.
trim	Upper percentile of data to trim.

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (survData, tra, type, standerr = FALSE, trim = 0.99) 
{
    tsurvData <- survData
    msm.data <- etmFormat(tsurvData, type)
    mvna.data <- mvna::mvna(data = msm.data, state.names = c("0", 
        "1", "2"), tra = tra, cens.name = "cens")
    etm.data <- etm::etm(data = msm.data, state.names = c("0", 
        "1", "2"), tra = tra, cens.name = "cens", s = 0)
    clos.data <- etm::clos(etm.data, aw = TRUE)
    if (standerr) {
        se <- sqrt(var(boot.clos(data = msm.data, state.names = c("0", 
            "1", "2"), tra = tra, cens.name = "cens", 0, nboot = 10)))
    }
    else {
        se <- 0
    }
    list(mvna.data = mvna.data, clos.data = clos.data, se = se, 
        etm.data = etm.data)
  }

n8thangreen/HESmanip documentation built on March 21, 2020, 12:20 a.m.