+ Reply to Thread
Results 1 to 3 of 3

Thread: Research methodology question

  1. #1
    Points: 8,411, Level: 61
    Level completed: 87%, Points required for next Level: 39

    Thanked 0 Times in 0 Posts

    Research methodology question

    We are implementing a new community healthcare program at work in order to reduce hospital attendances and admissions. The program began in June 2015 and the latest data we have is January 2016. During that time patients have been constantly referred to the program and hence the time in care can range from 0 to 7 months. Some patients have been discharged and are no longer receiving community care and some have been recently referred and awaiting assessment to see if they will be accepted into the care program.

    Management has come up with some KPIs for those currently on the program and those who have been discharged. Basically I’ve been told to just average the before and after attendances and admissions (before being 12 months prior to referral) as well as reporting the average length of time on the community program. I may be over thinking this but I think that is too simplistic and does not correctly convey what is actually happening and if we are seeing a true reduction.

    I was wondering if there was a methodology for analysing a dataset where observations are being added monthly and at any one time groups of patients have different lengths of time in care. If for example a patient had seven admissions before coming onto the program a month ago and now has zero admissions after one month, it doesn’t really say much, whereas in the same dataset a patient receiving community care for seven months with zero admissions compared to the twenty over the twelve months before, says a lot.

    Confused and hoping there is some existing methodology for this type of analysis. I'm not looking for answers, just pointers to methodologies for analysing this type of data which I can look up and research myself.

    Last edited by scotchpie; 03-17-2016 at 10:17 AM.

  2. #2
    TS Contributor
    Points: 14,811, Level: 78
    Level completed: 91%, Points required for next Level: 39
    Miner's Avatar
    Greater Milwaukee area
    Thanked 405 Times in 363 Posts

    Re: Research methodology question

    This is an "events in time" scenario. Events in time are typically analyzed using reliability techniques. Some of the reliability tools are referred to as survival analysis when you are in a non-product discipline, but the approach is similar.

    The terminology may sound confusing to a non-reliability person. For example, the use of failure. Just mentally replace the word failure with event, and the technique is the same. In your case, I would recommend a technique called reliability growth modeling. This is used to demonstrate whether the events in time are occurring at an increasing/decreasing/constant rate.

  3. #3
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Thanked 915 Times in 874 Posts

    Re: Research methodology question

    I think to start you need to clarify what the dependent variable is. And the predictors. I think, although I am not sure, that it is if you admitted to a hospital. And the predictors are if you attend some other program and (perhaps) how long. Then you can simply test if the percent difference for those that attended the hospitals for those who had and did not have the intervention differed significantly.

    But again I am not sure I have characterized either the predicted or dependent variables correctly.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ 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