Announcement

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

  • restricting time on cox model

    Hi, I wondered if anyone could help tell me which is the correct syntax here.

    I have a large data set where I want to risk, using a multivariate cox model, of surgery in people with an inflammatory bowel condition. I want to include people that only have at least 2 years of follow up and restrict the analysis to 2 years ie. look at what their risk of surgery was by 2 years of follow up from their time diagnosis.

    The two syntax I have are ;

    1.
    stset exit, origin(diagnosis_date1) failure(elective) scale(365.25) id(patid) enter(diagnosis_date1) exit(exit)
    stsplit twoyears, at(2)
    keep if twoyears == 2
    xi: stcox i.gender i.imd ib8.region i.diagnosis_age_cat i.charlson_fin i.diag_year_cat

    or


    2.
    stset exit, failure(elective) scale(365.25) origin(diagnosis_date) id(patid) exit(time diagnosis_date + 2*365.25)
    xi: stcox i.gender i.imd ib8.region i.diagnosis_age_cat i.charlson_fin i.diag_year_cat


    I found the second code on stata forum from someone who was looking at something similar but getting different results with each so if someone could explain which is correct I would be most grateful!

    Thanks
    Last edited by jennifer couch; 29 Jul 2024, 08:04.
Working...
X