Skip to contents

TableSubgroupCox: Sub-group analysis table for Cox/svycox model.

Source: R/forestcox.R
TableSubgroupCox.Rd

Sub-group analysis table for Cox/svycox model.

Usage

TableSubgroupCox(
  formula,
  var_subgroup = NULL,
  var_cov = NULL,
  data,
  time_eventrate = 3 * 365,
  decimal.hr = 2,
  decimal.percent = 1,
  decimal.pvalue = 3,
  cluster = NULL,
  strata = NULL,
  weights = NULL,
  event = FALSE,
  count_by = NULL,
  labeldata = NULL
)

Arguments

formula

formula with survival analysis.

var_subgroup

1 sub-group variable for analysis, Default: NULL

var_cov

Variables for additional adjust, Default: NULL

data

Data or svydesign in survey package.

time_eventrate

Time for kaplan-meier based event rate calculation, Default = 365 * 3

decimal.hr

Decimal for hazard ratio, Default: 2

decimal.percent

Decimal for percent, Default: 1

decimal.pvalue

Decimal for pvalue, Default: 3

cluster

Cluster variable for coxph, Default: NULL

strata

Strata variable for coxph, Default: NULL

weights

Weights variable for coxph, Default: NULL

event

Show number and rates of event in survival analysis default:F

count_by

Select variables to count by subgroup, Default: NULL

labeldata

Label info, made by `mk.lev` function, Default: NULL

Value

Sub-group analysis table.

Details

This result is used to make forestplot.

See also

safely,map,map2 coxph svycoxph confint

Examples

library(survival)
library(dplyr)
#> 
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#> 
#>     filter, lag
#> The following objects are masked from ‘package:base’:
#> 
#>     intersect, setdiff, setequal, union
lung %>%
  mutate(
    status = as.integer(status == 1),
    sex = factor(sex),
    kk = factor(as.integer(pat.karno >= 70))
  ) -> lung
TableSubgroupCox(Surv(time, status) ~ sex, data = lung, time_eventrate = 100)
#>      Variable Count Percent Point Estimate Lower Upper sex=1 sex=2 P value
#> sex2  Overall   228     100           1.91  1.14   3.2     0   1.2   0.014
#>      P for interaction
#> sex2                NA
TableSubgroupCox(Surv(time, status) ~ sex,
  var_subgroup = "kk", data = lung,
  time_eventrate = 100
)
#>   Variable Count Percent Point Estimate Lower Upper sex=1 sex=2 P value
#> 1       kk  <NA>    <NA>           <NA>  <NA>  <NA>  <NA>  <NA>    <NA>
#> 2        0    38    16.9           2.88  0.31 26.49     0     0    0.35
#> 3        1   187    83.1           1.84  1.08  3.14     0   1.5   0.026
#>   P for interaction
#> 1             0.525
#> 2              <NA>
#> 3              <NA>

## survey design
library(survey)
data.design <- svydesign(id = ~1, data = lung)
#> Warning: No weights or probabilities supplied, assuming equal probability
TableSubgroupCox(Surv(time, status) ~ sex, data = data.design, time_eventrate = 100)
#> Independent Sampling design (with replacement)
#> svydesign(id = ~1, data = lung)
#>      Variable Count Percent Point Estimate Lower Upper sex=1 sex=2 P value
#> sex2  Overall   228     100           1.91  1.14  3.19     0   1.2   0.013
#>      P for interaction
#> sex2                NA
TableSubgroupCox(Surv(time, status) ~ sex,
  var_subgroup = "kk", data = data.design,
  time_eventrate = 100
)
#> Independent Sampling design (with replacement)
#> svydesign(id = ~1, data = lung)
#> New names:
#> • `` -> `...1`
#> • `` -> `...2`
#> Independent Sampling design (with replacement)
#> subset(data, get(var_subgroup) == .)
#> Independent Sampling design (with replacement)
#> subset(data, get(var_subgroup) == .)
#>   Variable Count Percent Point Estimate Lower Upper sex=1 sex=2 P value
#> 1       kk  <NA>    <NA>           <NA>  <NA>  <NA>  <NA>  <NA>    <NA>
#> 2        0    38    16.9           2.88  0.31  27.1     0     0   0.355
#> 3        1   187    83.1           1.84  1.08  3.11     0   2.9   0.024
#>   P for interaction
#> 1             0.523
#> 2              <NA>
#> 3              <NA>