conf.level in survdiff_ci() function

[mkocherg]

Hi,

I don't think the conf.level option works as intended in the survdiff_ci() function. When I change it, the conf intervals remain the same, but SE and pvalue change. Could you please take a look?

data(cancer, package = "survival")

cancer <- cancer |>
    dplyr::mutate(
        sex = factor(
            sex,
            levels = 1:2,
            labels = c("Male", "Female")
        ),
        status = status - 1
    )


survdiff_ci(
    formula = survival::Surv(time = time, event = status) ~ sex,
    data = cancer,
    time = 365.25
)
survdiff_ci(
    formula = survival::Surv(time = time, event = status) ~ sex,
    data = cancer,
    time = 365.25, conf.level = .9
)

[stopsack]

Hi Masha- Thanks a lot for your but report and the reprex. I see how the confidence level argument is not passed on in one of the function calls within survdiff_ci(). Please give me a few days to add a pull request. Meanwhile, if you are interested in estimating differences (as opposed to ratios), you can switch to approach = "squareadd", which is not affected and gives reasonable estimates in general as well.

[mkocherg]

Thank you so much! Happy new year!

…

On Fri, Jan 2, 2026 at 9:38 AM Konrad H. Stopsack ***@***.***> wrote: *stopsack* left a comment (stopsack/rifttable#8) <#8 (comment)> Hi Masha- Thanks a lot for your but report and the reprex. I see how the confidence level argument is not passed on in one of the function calls within survdiff_ci(). Please give me a few days to add a pull request. Meanwhile, if you are interested in estimating differences (as opposed to ratios), you can switch to approach = "squareadd", which is not affected and gives reasonable estimates in general as well. — Reply to this email directly, view it on GitHub <#8 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AB7YGE32U4J6YASZSYETSDD4E2GHNAVCNFSM6AAAAACQLVBAZ2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTOMBVGYYDINRZHA> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[stopsack]

@mkocherg - the bug should be fixed now, see PR #9. The update is available in the dev branch already, installable via remotes::install_github("stopsack/rifttable@dev") and will be released in v0.7.2 shortly.

[stopsack]

v0.7.2. is on CRAN