Competing risk analysis

Jinseob Kim

2024-12-13

Display results of comepting risk analysis using jstable(Fine-Gray Method)

TableSubgroupMultiCox

When using the TableSubgroupMultiCox function, preprocessing the data with the finegray function from the survival package is required. The finegray function generates a new dataset containing fgstart, fgstop, fgstatus, and fgwt. The TableSubgroupMultiCox function then displays results based on the corresponding formula and weights.

data <- mgus2
data$etime <- with(data, ifelse(pstat == 0, futime, ptime))
data$event <- with(data, ifelse(pstat == 0, 2 * death, 1))
data$event <- factor(data$event, 0:2, labels = c("censor", "pcm", "death"))
data$age65 <- with(data, ifelse(age > 65, 1, 0))
data$age65 <- factor(data$age65)
pdata <- survival::finegray(survival::Surv(etime, event) ~ ., data = data)
TableSubgroupMultiCox(formula = Surv(fgstart, fgstop, fgstatus) ~ sex, data = pdata, var_cov = "age", weights = "fgwt", var_subgroups = c("age65"))
#>     Variable Count Percent Point Estimate Lower Upper sex=F sex=M P value
#> sex  Overall 41775     100           0.77  0.54   1.1  10.4   8.1   0.153
#> 1      age65  <NA>    <NA>           <NA>  <NA>  <NA>  <NA>  <NA>    <NA>
#> 2          0  7572    18.1           0.71   0.4  1.27  17.2  10.5   0.252
#> 3          1 34203    81.9            0.8  0.51  1.25   8.7   7.3    0.33
#>     P for interaction
#> sex              <NA>
#> 1               0.673
#> 2                <NA>
#> 3                <NA>

cox2.display

As written above, preprocessing the data with finegray function is also required. By using corresponding formula and weights, cox2.display function will display table results.

fgfit <- coxph(Surv(fgstart, fgstop, fgstatus) ~ age + sex,
  weight = fgwt, data = pdata, model = T
)
cox2.display(fgfit)
#> $table
#>             crude HR(95%CI)    crude P value adj. HR(95%CI)     adj. P value
#> age         "0.98 (0.97,0.99)" "0.003"       "0.98 (0.97,0.99)" "0.002"     
#> sex: M vs F "0.8 (0.56,1.13)"  "0.207"       "0.77 (0.54,1.1)"  "0.153"     
#> 
#> $metric
#>                          [,1] [,2] [,3] [,4]
#> <NA>                       NA   NA   NA   NA
#> No. of observations 41775.000   NA   NA   NA
#> No. of events         115.000   NA   NA   NA
#> AIC                  1583.872   NA   NA   NA
#> 
#> $caption
#> [1] "Cox model on time ('fgstart, fgstop') to event ('fgstatus')"