+ Reply to Thread
Results 1 to 6 of 6

Thread: Please help me select the correct statistical test

  1. #1
    Points: 11, Level: 1
    Level completed: 21%, Points required for next Level: 39

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Please help me select the correct statistical test




    I'm a general surgery resident and am conducting a quality improvement study about waste in the OR. My statistics background is limited to what I was taught during medical school and what I've taught myself over the years while conducting research. I'm using Stata 12 to perform my analysis.

    In simple, my study is examining surgical waste in the OR due to setup, nursing staff, and surgeons with and without an intervention to decrease waste. Unfortunately it doesn't appear that the intervention was effective, but there does seem to be a difference in number of wasted items between the three groups (setup, nursing, surgeon). I'm trying to determine the best test to compare the three groups. I essentially need a test of three continuous variables. I believe it is an ANOVA but I'm not sure and also don't know exactly how to code this in Stata. An example of my database is as such:

    Observation; Pre/Post intervention; Waste(setup); Waste(nurse); Waste(surgeon)
    1 Pre 3 3 0
    2 Pre 5 2 1
    3 Pre 2 1 0
    4 Pre 7 2 1
    5 Pre 8 0 2
    6 Post 7 2 1
    7 Post 6 1 1
    8 Post 3 0 0
    9 Post 2 1 0
    10 Post 5 3 2

    This is just an example of the database's formatting and not representative of the actual data. Please let me know the best way to compare the three groups. As an added request, I would greatly appreciate any advice on how to code it in Stata but if not I can figure that out on my own if I know which test to use. What I'm hoping to be able to conclude from this dataset would be simply that most surgical waste is due to setup and the interventions implemented were ineffective and other means to decrease waste could be such and such and such directed at blah blah blah as evidenced by blah.

    Thank you in advance for any help!

  2. #2
    Omega Contributor
    Points: 38,396, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,001
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Please help me select the correct statistical test

    What is the outcome variable in particular?

    What year are you? Glad to see your program endorsing QI.
    Stop cowardice, ban guns!

  3. #3
    Points: 11, Level: 1
    Level completed: 21%, Points required for next Level: 39

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Please help me select the correct statistical test

    I'm an intern. Most of this is self-directed research since I'm at a county program and most individuals who do research do it exclusively during elective research years. This is also why I don't have many resources to assist me in statistical analysis within my program.

    I have a few experimental variables but the main one is Pre/Post intervention (others include surgical length, surgeon, type of surgery, first start; I've made these all bivariate) and my main outcome variable is the continuous variable "total items wasted". I'm just doing t-tests for most of the analysis because the experimental variables are bivariate. However, I sub-classified wasted items as being wasted due to three different groups (setup/nursing/surgeon) and wanted a way beyond a T-test of A vs. B, A vs. C, and B vs. C to compare the mean number of items wasted between these groups. As I mentioned my experimental variables didn't seem to have much effect so it's not crucial that this be included in the test of means but if there is a way to include it I would like to.

    Waste seems to be predominantly due to setup (i.e items being opened but not used prior to the start of the case) and there are interventions I have in mind to decrease this, but I need a statistical test that can identify and say that yes the waste is significantly higher due to setup than intra-operative requests so I can focus my discussion on interventions directed at reducing this source of waste.

    Most of my prior research was epidemiological in infectious disease and I was able to write up the majority of my papers using chi square, wilcoxon-rank sum, multivariable logistic regressions, and post-hoc analyses so I'm not that familiar at working with multiple continuous variables. Realistically, I probably should just pursue an MPH in biostats during a research year haha but for now youtube and resources like this is what I'm working with.

  4. #4
    Omega Contributor
    Points: 38,396, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,001
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Please help me select the correct statistical test

    What I was probing at in my question was that your outcome seems as though it is a count not a continuous variable.
    Stop cowardice, ban guns!

  5. #5
    Points: 11, Level: 1
    Level completed: 21%, Points required for next Level: 39

    Posts
    3
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Please help me select the correct statistical test

    I'm treating each case as an individual observation so I believe it's continuous since I'm comparing means and it has a relatively normal distribution, but I'm not that familiar with counts or statistical tests to analyze them.

  6. #6
    TS Contributor
    Points: 12,227, Level: 72
    Level completed: 45%, Points required for next Level: 223
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,470
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: Please help me select the correct statistical test


    hi,
    as a lean consultant, I would advise a different approach. Just make a list of the wastes you think are important a decide on priorities for them. This decision should involve several factors and will be subjective :e.g. based on amount of waste but also ease of eliminating it, impact on the operation etc. Then pick one and work on it without worrying whether it is significantly higher than the others or not.

    good luck

+ 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