cox_output: Calculate cox regression and extract Hazard Ratio and pvalues

In MBender1992/emR: Dictionary for EuMelaReg functions (analysis of clinical studies)

Calculate cox regression and extract Hazard Ratio and pvalues

Description

This function calculates a cox regression with the coxph function of the survival package (univariate if number of vars = 1, multivariate if number of vars > 1). Hazard ratios and pvalues are extracted and shown in a table format. If a variable is divided in more than 2 groups the wald test statistic for overall significance is also shown. This function calculates a univariate cox regression with the coxph function from the survival package if the argument vars is length 1; otherwise a multivariate cox regression is calculated. Hazard ratios and pvalues are extracted and shown in a table format. If a variable in univariate cox regression is divided in more than 2 groups the wald test statistic for overall significance is also shown.

Usage

cox_output(
  data,
  time,
  status,
  vars,
  fixed.var = NULL,
  output = "table",
  modeltype = "full",
  p.thres = 0.1,
  niter = 10,
  weights = NULL
)

Arguments

data	data.frame or data.table containing survival data.
time	the time interval from start of observation until date of event (e.g. disease progression or death) or censoring.
status	variable specifying if event occured or data has been censored.
vars	variables tested for Influence on outcome. Defines type of cox regression: if length is 1 the output is a univariate cox regression, for length > 1 the output is a multivariate cox regression
fixed.var	specifies fixed variables to be included in the cox model.
output	Defines the output object of the function. Default is \"table\". Another possible option is \"fit\" to return a coxph object.
modeltype	character value. Allowed values include: "full" and "backwards". Specify if full model or backwards selection model should be used for multivariate cox regression.
p.thres	pvalue threshold for backwards selection model.
niter	number of iterations for backwards selection model.
weights	character variable specifying the name of the weights column. Weights have to be added to the original dataframe in order to be applied correctly.

MBender1992/emR documentation built on Feb. 18, 2025, 9:21 a.m.