+ Reply to Thread
Results 1 to 2 of 2

Thread: getting in deep..

  1. #1
    Points: 30, Level: 1
    Level completed: 60%, Points required for next Level: 20

    Thanked 0 Times in 0 Posts

    getting in deep..


    I am attempting to do statistical analysis on data that I've collected, but I am starting to feel overwhelmed by the choices as I read through the posts on the board. The data consists of approximately 12 data points measured by echocardiography (ultrasound of the heart). I am trying to measure the change on the diastology of the heart after a TAVR valve has been deployed. There are up to five instances that these data points are measured ( pre-procedure, discharge, 30 day, 1 y and 2 y). In some instances, not all of the data points are reported so there are gaps or the patient passes on. My end goal is to find a cluster of patients that improve the fastest based off of the data and identify what factor(s) are present that allow them to do so. My theory is that those factors will be identified through a rate of change, but I am unable to articulate that in a statistical fashion. If there is additional information that you would need to form a proper response to help be more efficiently, please don't hesitate to ask. I've attached an excel spreadsheet as an example of what the data collection looks like so far. Thank you ahead of time for your help.
    Attached Files

  2. #2
    TS Contributor
    Points: 17,749, Level: 84
    Level completed: 80%, Points required for next Level: 101
    Karabiner's Avatar
    FC Schalke 04, Germany
    Thanked 640 Times in 602 Posts

    Re: getting in deep..

    Is it possble to define "response"? E.g. something like
    "response = absolute change from baseline > xyz", or
    "response = change from baseline > XYZ%". You could
    then use survival analysis with response as terminal

    With kind regards


+ Reply to 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