[Feature Request]: should ard_survival_survfit() have a conf.level argument?

[larmarange]

Feature description

Currently, only 95% of confidence intervals are returned. Customizing the confidence interval may be relevant.

Code of Conduct

• I agree to follow this project's Code of Conduct.

Contribution Guidelines

• I agree to follow this project's Contribution Guidelines.

Security Policy

• I agree to follow this project's Security Policy.

[ddsjoberg]

Thanks @larmarange ! Users can control the confidence level by specifying it in the survfit(conf.int) call. BUT we should make this mode clear by adding an example with a different conf level and I also think we should return the conf.level with the results.

library(cardx)

survival::survfit(ggsurvfit::Surv_CNSR(AVAL, CNSR) ~ 1, cards::ADTTE, conf.int = 0.80) |>
  ard_survival_survfit(times = 60) 
#> {cards} data frame: 5 x 9
#>   variable variable_level  context stat_name stat_label  stat
#> 1     time             60 survival    n.risk  Number o…    93
#> 2     time             60 survival  estimate  Survival… 0.457
#> 3     time             60 survival std.error  Standard… 0.033
#> 4     time             60 survival conf.high  CI Upper… 0.502
#> 5     time             60 survival  conf.low  CI Lower… 0.416
#> ℹ 3 more variables: fmt_fn, warning, error

survival::survfit(ggsurvfit::Surv_CNSR(AVAL, CNSR) ~ 1, cards::ADTTE, conf.int = 0.90) |>
  ard_survival_survfit(times = 60) 
#> {cards} data frame: 5 x 9
#>   variable variable_level  context stat_name stat_label  stat
#> 1     time             60 survival    n.risk  Number o…    93
#> 2     time             60 survival  estimate  Survival… 0.457
#> 3     time             60 survival std.error  Standard… 0.033
#> 4     time             60 survival conf.high  CI Upper… 0.515
#> 5     time             60 survival  conf.low  CI Lower… 0.405
#> ℹ 3 more variables: fmt_fn, warning, error

^{Created on 2024-09-26 with reprex v2.1.1}

[ddsjoberg]

I went to add conf.level to the returned results, but I wasn't sure exactly what it should look like ... so just updated the example instead. #218

[larmarange]

thanks