Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • artbin: Extended sample size for randomised trials with binary outcomes

    Dear Stata List

    I am very happy to announce a major update to the user-written command artbin, version 2.0, now available via SSC.

    artbin can be used to calculate sample size and power for trials with a binary outcome variable, including non-inferiority trials.

    It is part of the Analysis of Resources for Trials package (ART) on SSC, available by typing:

    ssc install art

    artbin was first written in 2004 by Prof Abdel Babiker and Prof Patrick Royston, with contribution from Sophie Barthel. It has recently undergone a major update with invaluable input from Prof Ian White, Prof Ab Babiker, Prof Patrick Royston, Prof Max Parmar and my humble non-Prof self. I presented the package and it’s updates at the recent 2021 ‘London’ UK Stata Conference.

    One of the main features of artbin that sets it apart from the other available software in Stata is the range of trial types, statistical tests and methods that it offers for sample size calculation.

    All of the exciting features in artbin, and differences from the previous version are fully documented in
    artbin_whatsnew.sthlp. In particular:


    • A more flexible approach to handling non-inferiority trials is enabled, with a margin(#) option now included.
    • The wald option has been included for the Wald test.
    • The user can now specify whether the outcome is favourable or unfavourable for a 2-arm trial.
    • Sample size per group is now reported, and rounding up to the nearest integer is performed per group.
    • A noround option has been included for the case when the user does not want artbin to round the calculated sample size.
    • A loss to follow-up option is now available (ltfu(#)).

    The user will need to alter previous coding using artbin to accommodate the following above changes, as detailed in the
    *.sthlp files.
    With thanks as ever to Kit Baum for kindly updating SSC with our new artbin package.

    Please don't hesitate to contact me ([email protected]) with any queries or comments.

    I hope you find the updated package useful.

    Many thanks and all the very best

    Ella


    Ella Marley-Zagar
    Sr Research Associate
    MRC Clinical Trials Unit at UCL, London, UK



  • #2
    Dear users of artbin,

    An updated version of artbin (v2.1) was posted on SSC yesterday, with the following main changes:
    • artbin had a major error when using the ltfu() option and calculating power from sample size. This has now been corrected.
    • artbin had a minor error in rounding of sample sizes, which did not behave exactly as advertised. This has also been corrected: rounding is now done as stated in the help file, i.e. sample sizes are rounded up to the nearest integer in each group. Sample sizes may now differ by 1 or 2 from previous calculations.
    Thank you.

    Ian

    Comment

    Working...
    X