In SandraCastroPearson/BivRec: Bivariate Alternating Recurrent Event Data Analysis

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-"
)

Alternating recurrent event data arise frequently in biomedical and social sciences where two types of events such as hospital admissions and discharges occur alternatively over time. BivRec implements a collection of nonparametric and semiparametric methods to analyze such data.

The main functions are:
- bivrecReg: Use for the estimation of covariate effects on the two alternating event gap times (Xij and Yij) using semiparametric methods. The method options are "Lee.et.al" and "Chang".
- bivrecNP: Use for the estimation of the joint cumulative distribution function (cdf) for the two alternating events gap times (Xij and Yij) as well as the marginal survival function for type I gap times (Xij) and the conditional cdf of the type II gap times (Yij) given an interval of type I gap times (Xij) in a nonparametric fashion.

The package also provides options to simulate and visualize the data and results of analysis.

Installation

BivRec depends on the following system requirements:
- Rtools. Download Rtools 35 from https://cran.r-project.org/bin/windows/Rtools/

Once those requirements are met you can install BivRec from github as follows:

#Installation requires devtools package.
#install.packages("devtools")
library(devtools)
#install_github("SandraCastroPearson/BivRec")

Example

This is an example using a simulated data set.

# Simulate bivariate alternating recurrent event data
library(BivRec)
set.seed(28)
sim_data <- simBivRec(nsize=100, beta1=c(0.5,0.5), beta2=c(0,-0.5), tau_c=63, set=1.1)
head(sim_data)

# Create a bivrecSurv object
bivrec_object <- with(sim_data, bivrecSurv(id, epi, xij, yij, d1, d2))
# Plot gap times
plot(bivrec_object, main="Example", type = c("Type I", "Type II"), col = c("darkgrey","darkgreen"))

Nonparametric Analysis

# Apply the nonparametric method of Huang and Wang (2005) and visualize joint, marginal and conditional results

library(BivRec)
npresult <- bivrecNP(response = with(sim_data, bivrecSurv(id, epi, xij, yij, d1, d2)),
                ai=1, u1 = seq(1, 15, 0.5), u2 = seq(1, 15, 0.5), conditional = TRUE,
                given.interval = c(1, 10), level = 0.99)
head(npresult)
plot(npresult)

# To save individual plots in a pdf file un-comment the following line of code: 
# pdf("nonparam_jointcdfplot.pdf")
# plotJoint(npresult)
# dev.off()
# pdf("nonparam_marginalplot.pdf")
# plotMarg(npresult)
# dev.off()
# pdf("nonparam_conditionaplot.pdf")
# plotCond(npresult)
# dev.off()

Semiparametric Regression Analysis

#Explore how the response changes by levels of a categorical covariate using a plot. 
#Use attach as follows or specifiy each vector using the $ operator (sim_data$id, sim_data$epi, etc.)

attach(sim_data)
plot(x = bivrecSurv(id, epi, xij, yij, d1, d2), by = data.frame(a1, a2), 
     type = c("Type I", "Type II"), col = c("darkgrey","darkgreen"))
detach(sim_data)

# Apply Lee, Huang, Xu, Luo (2018) method using multiple covariates.
lee_fit <- bivrecReg(bivrecSurv(id, epi, xij, yij, d1, d2) ~ a1 + a2,
                    data= sim_data, "Lee.et.al")

summary(lee_fit)

# To apply Chang (2004) method use method="Chang".
# chang_fit <- bivrecReg(bivrecSurv(id, epi, xij, yij, d1, d2) ~ a1 + a2,
#                    data= sim_data, "Chang")