Discussion: View Thread

  • 1.  Stratification

    Posted 03-16-2022 23:39
    Recently I reviewed a protocol with the phrase "Patients will be stratified by gender."
    My initial reaction was that this is incorrect and it should be "Randomization will be stratified by gender."
    The phrase is used several times throughout the protocol and SAP, so I didn't comment on it.
    Is their wording correct? Should I have noted my opinion? Am I being picky (or wrong)?
    Thanks
    David

    ------------------------------
    David Bristol
    Statistical Consulting Services, Inc.
    ------------------------------


  • 2.  RE: Stratification

    Posted 03-17-2022 00:17
    Not picky. I'd request the change.  The wording will live on to the final clinical study report and possibly a publication(s), press releases etc..

    ------------------------------
    Chris Barker, Ph.D.
    2022 Statistical Consulting Section
    Chair-elect
    Consultant and
    Adjunct Associate Professor of Biostatistics
    www.barkerstats.com


    ---
    "In composition you have all the time you want to decide what to say in 15 seconds, in improvisation you have 15 seconds."
    -Steve Lacy
    ------------------------------

    Original Message


  • 3.  RE: Stratification

    Posted 03-17-2022 00:50
    Agree with Chris Barker. It's our job as statisticians to keep the language as precise as we can in the Protocol, and especially the SAP. Good highlight!

    ------------------------------
    -
    Eric J. Daza, DrPH, MPS 🇺🇸🇵🇭 (he/him) | ericjdaza.com | ericjdaza@gmail.com
    Creator + Editor at Stats-of-1 | statsof1.org | ericjdaza@statsof1.org
    Lead Biostatistician in Data Science at Evidation Health | evidation.com | edaza@evidation.com
    ------------------------------

    Original Message


  • 4.  RE: Stratification

    Posted 03-17-2022 03:10
    May be ok, if context is clear. If subheaders "Randomization" and "Analysis" are followed by "Patients will be stratified by gender", that is nearly as clear as the preferable sentences "Randomization will be stratified by gender" and "Analysis will be stratified by gender".

    ------------------------------
    Reinhard Vonthein
    ------------------------------

    Original Message


  • 5.  RE: Stratification

    Posted 03-17-2022 10:54
    Agree, the trouble is really that the context isn't clear; I can think of at least three things that could be meant:
    1. stratified sampling by gender
    2. stratified randomization of a treatment within gender
    3. an analysis that will include gender as a covariate
    And of course it could mean more than one of these, or even all three.

    ------------------------------
    Aaron Rendahl, Ph.D.
    Assistant Professor of Statistics and Informatics
    College of Veterinary Medicine, University of Minnesota
    300A VetSci, 612-301-2161
    ------------------------------

    Original Message


  • 6.  RE: Stratification

    Posted 03-17-2022 16:48
    Trials that say stratified by gender are generally powered for each gender.  This is required for many NIH trials.

    Not only is there an overall analysis that corrects for imbalance of treatment arms between genders and looks for treatment by gander interaction, gender-specific point and interval estimates of effect size is expected.

    ------------------------------
    Jonathan Shuster
    ------------------------------

    Original Message


  • 7.  RE: Stratification

    Posted 03-17-2022 17:03
    Thanks for pointing out this inconsistency of usage of the term. What you describe would be called subgroup analysis in other clinical trials.

    ------------------------------
    Reinhard Vonthein
    ------------------------------

    Original Message


  • 8.  RE: Stratification

    Posted 03-20-2022 12:03
    I agree here with David Bristol, Chris Barker, & Eric Daza. Yet the even bigger mistake is using the word "gender" here. Presumably they mean "Randomization will be stratified by sex."

    http://www.icmje.org/icmje-recommendations.pdf


    ------------------------------
    Michael Hanna, PhD
    Mercury Medical Research & Writing
    www.mercury-mrw.com


    https://link.springer.com/book/10.1007/978-3-030-02955-5#about
    ------------------------------

    Original Message