+ Reply to Thread
Results 1 to 3 of 3

Thread: Help with kaplan meier analysis in SPSS

  1. #1
    Points: 380, Level: 7
    Level completed: 60%, Points required for next Level: 20

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Help with kaplan meier analysis in SPSS




    HI!

    Thank you very much in advance for helping with my post!

    I am a phd student, I do surgical research at the carotid artery. My current project looks at plaque echolucency, which is a radiological finding on ultrasound examination. Your carotid artery can either be echolucent (1) or non-echolucent (2)

    We followed up a cohort of patients with a carotid artery plaque for 10 years. At baseline the plaque could be echolucent (1) or non-echolucent (2). I am interested in the risk of stroke in the 2 groups.
    In the follow up of 10 years one of the following could happen to a patient. stroke, death, lost to follow up, surgery or end of study time (10 years) without one of the previous scenarios.

    I have performed a kaplan meier analysis by creating 3 columns in SPSS
    1. serial time = defined by time at beginning of study until one of the following dates: time to event, time to death, lost to FU, ipsilateral surgery, or end of study time so 10 years
    2. Status = status at end of serial time: event (=stroke) or no event (=death, lost to FU, ipsilateral surgery, end of study time without one of the previous scenarios)
    3. group = echolucent (1) or non echolucent (2)

    My question are as follows:
    1. Did I correctly define the status in column 2?
    2. How do I find the 5 and 10 year cum. stroke risk (with 95% CI) in SPSS for echolucent and non-echolucent plaques?
    3. The overall comparison box gives the following values:
    Log rank (Mantel Cox): Chi-square 0.777; df 1; Sig 0.378.
    Does this mean that overall plaque echolucency does not have a significant influence
    stroke risk?
    4. How can I compare if the 5 year cum stroke risk is sign. different between both groups?

    Thanks again! I could really use your help, please let me know if anything is unclear!

  2. #2
    TS Contributor
    Points: 17,779, Level: 84
    Level completed: 86%, Points required for next Level: 71
    Karabiner's Avatar
    Location
    FC Schalke 04, Germany
    Posts
    2,542
    Thanks
    56
    Thanked 640 Times in 602 Posts

    Re: Help with kaplan meier analysis in SPSS

    1. Did I correctly define the status in column 2?
    Seemingly.
    2. How do I find the 5 and 10 year cum. stroke risk (with 95% CI) in SPSS for echolucent and non-echolucent plaques?
    You can ask for an output table within the
    Kaplan-Meier procedure where cum. survival
    is listed (hopefully, with standard error).
    3. The overall comparison box gives the following values:
    Log rank (Mantel Cox): Chi-square 0.777; df 1; Sig 0.378.
    Does this mean that overall plaque echolucency does not have a significant influence
    stroke risk?
    There was no significant difference regarding "survival times"
    (until stroke), yes.
    Interpretation could become difficult if death/sugery/loss
    are distributed markedly different between groups, or if groups
    showed markedly different stroke-related characteristics at
    baseline.
    How can I compare if the 5 year cum stroke risk is sign. different between both groups?
    You could create a column with status at year 5
    and actual observation time at year 5, according
    to what you did with regard to 10 years.

    With kind regards

    K.

  3. #3
    Points: 380, Level: 7
    Level completed: 60%, Points required for next Level: 20

    Posts
    4
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Help with kaplan meier analysis in SPSS


    Thank you so much for your help. That's great!

    In addition to this, perhaps you can help me with the following:

    I also have data on there follow up scans within the 10 years (not all of them had a scan every year though). I would like to know if an increase in plaque echolucency is related to risk of stroke.

    I have the following data for each subject.
    subject 1: follow up 1 (increase echo lucency: no); follow up 2 (increase echolucency yes); follow up 3 (increase echolucency yes)
    subject 2: follow up 1 (increase echo lucency yes); follow up 2 (increase echo lucency yes)
    … and this for a few hundred subjects.

    Could you advise me on a way to look at the association between increase in plaque echo lucency and stroke risk with the data i have available?

    KR

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats