Interval-Censored Cure Models

Performs cure model analysis for interval-censored survival data using the ICGOR (Interval-Censored Generalized Odds Rate) methodology. This approach is specifically designed for situations where the exact event time is not known, but falls within an interval, and a fraction of the population may be cured (i.e., will never experience the event).

Usage

intervalcensorcure(
  data,
  leftTime,
  rightTime,
  covariates,
  survivalDistribution = "weibull",
  cureModel = "mixture",
  includeUncured = TRUE,
  confidenceLevel = 0.95,
  maxIterations = 1000,
  tolerance = 1e-06,
  bootstrapSamples = 500,
  performBootstrap = FALSE,
  modelComparison = TRUE,
  plotSurvival = TRUE,
  plotCure = TRUE
)

Arguments

data

the data as a data frame

leftTime

the left endpoint of the interval for interval-censored observations

rightTime

the right endpoint of the interval for interval-censored observations

covariates

covariates to include in the cure model

survivalDistribution

the parametric survival distribution for the uncured fraction

cureModel

the type of cure model to fit

includeUncured

whether to include detailed analysis of the uncured fraction

confidenceLevel

the confidence level for parameter estimates

maxIterations

maximum number of iterations for model convergence

tolerance

convergence tolerance for parameter estimation

bootstrapSamples

number of bootstrap samples for confidence intervals

performBootstrap

whether to perform bootstrap analysis for confidence intervals

modelComparison

whether to perform model comparison across different distributions

plotSurvival

whether to plot estimated survival functions

plotCure

whether to plot cure fraction estimates

Value

A results object containing:

results$instructions					a html
results$modelSummary					a table
results$cureResults					a table
results$survivalResults					a table
results$modelFit					a table
results$modelComparison					a table
results$bootstrapResults					a table
results$survivalPlot					an image
results$curePlot					an image
results$modelSummaryText					a html

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$modelSummary$asDF

as.data.frame(results$modelSummary)

Details

Key features:

• Mixture and non-mixture cure models

• Various survival distributions (Weibull, exponential, log-normal, log-logistic)

• Covariate effects on both cure fraction and survival parameters

• Model comparison and goodness-of-fit assessment

• Bootstrap confidence intervals

• Parametric and non-parametric approaches

Examples

# Interval-censored cure model with Weibull distribution
intervalcensorcure(
    data = data,
    leftTime = "left_time",
    rightTime = "right_time",
    covariates = c("age", "treatment"),
    survivalDistribution = "weibull",
    cureModel = "mixture",
    includeUncured = TRUE
)