Results 1 to 3 of 3

Thread: How to best visually represent rate of loss data where participants lost to followup?

  1. #1
    Newbie
    Joined
    Apr 2017
    From
    new zealand
    Posts
    2

    Exclamation How to best visually represent rate of loss data where participants lost to followup?

    Hi,

    I am currently doing a hospital audit on the rate of loss of immunity to Hepatitis B in dialysis patients. This involves identifying 100 patients who are hepatitis B immune, and following them up every 6 months with a blood test to confirm persisting immunity, over 5 years. Due to the high mortality rate of dialysis patients, there is a significant number of patients that are lost to followup over this 5 year period (around 40). Some of these patients remain hepatitis B immune up until they are lost to followup, while others become non-immune and are then lost to follow-up.

    How can I best display this data, taking into account those that are lost to followup, without affecting the results?

    Thanks heaps
    Follow Math Help Forum on Facebook and Google+

  2. #2
    Senior Member
    Joined
    Mar 2008
    From
    Pennsylvania, USA
    Posts
    280
    Thanks
    37

    Re: How to best visually represent rate of loss data where participants lost to follo

    I'm guessing you are performing a survival analysis, in which case, you could treat the deceased as withdraws from your study, though that may or may not introduce an undesirable amount of bias. There are a bevy of methods and tools available to "clean up" incomplete (i.e., truncated or censored) data, but I cannot get any more specific without you getting more specific. And what do you mean by "display" this data? Good luck!
    Follow Math Help Forum on Facebook and Google+

  3. #3
    Newbie
    Joined
    Apr 2017
    From
    new zealand
    Posts
    2

    Re: How to best visually represent rate of loss data where participants lost to follo

    Hi

    Thanks so much for your reply. I have tried analysing the data by excluding those lost to followup (survival analysis), but more often these are hepatitis B positive patients, meaning the the percentage of remaining positive patients is too low, and thus exaggerating the rate of loss (bias is introduced).

    I calculate the proportion that remain hepatitis b positive at each 6 month check point as follows:
    Total number that remain positive/total number of people remaining

    I define the total number of people remaining as the initial study population minus those lost to followup at that point in time.

    When i said 'display' data I meant in a graphical/pictorial format.

    Please let me know what specific information would be useful to advise on methods to 'clean up' this incomplete data.

    Thanks in advance
    Follow Math Help Forum on Facebook and Google+

Similar Math Help Forum Discussions

  1. 3 related rate problems, I am truly lost.
    Posted in the Calculus Forum
    Replies: 3
    Last Post: Oct 21st 2012, 07:55 PM
  2. Trying to find a function to represent my data
    Posted in the Calculus Forum
    Replies: 3
    Last Post: Aug 9th 2012, 03:11 PM
  3. Write an equation to represent the data.
    Posted in the Algebra Forum
    Replies: 3
    Last Post: Apr 30th 2011, 01:27 AM
  4. How do I best represent this data?
    Posted in the Math Topics Forum
    Replies: 1
    Last Post: Nov 19th 2010, 03:59 AM
  5. Back calculate total participants from rate ratios
    Posted in the Advanced Statistics Forum
    Replies: 0
    Last Post: Sep 29th 2010, 12:48 PM

/mathhelpforum @mathhelpforum