+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 15 of 22

Thread: Pool data and compare data for my systematic review

  1. #1
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Post Pool data and compare data for my systematic review/meta-analysis




    Hello everybody,

    I am new to this website and to statistics as well. Currently I am writing a systematic review on complications and reoperations after distal radius fractures(Systematic review on the comparison of complications and reoperations after plaster casting, K-wire fixation, Kapandji intrafocal pinning, external fixation, dorsal plating, volar plating and intramedullary fixation of extra-articular distal radius fractures in adults).

    I collected all the data and got stuck on the statistics. I tried to look for it in books, Google and YouTube but couldn't find the necessary information.

    How do I pool the data shown below? And secondly how do I compare the pooled data from different treatment modalities? Your help would be highly appreciated!!!!

    Study------ EF complications ---EF total study population--Compl Rate
    1------------3-----------------------23-----------------------------0,130434783
    2------------2-----------------------20-----------------------------0,1
    3------------ 8-----------------------30-----------------------------0,266666667
    4------------ 1-----------------------16-----------------------------0,0625
    5------------ 31---------------------39-----------------------------0,794871795
    6------------ 26---------------------75-----------------------------0,346666667
    7------------ 22---------------------75-----------------------------0,293333333
    8------------ 0-----------------------20 -----------------------------0

    Total-------93----------------------298---------------------------0,312080537
    Last edited by Rotterdam; 01-27-2017 at 11:49 AM.

  2. #2
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Pool data and compare data for my systematic review

    Quote Originally Posted by Rotterdam View Post
    Hello everybody,

    I am new to this website and to statistics as well. Currently I am writing a systematic review on complications and reoperations after distal radius fractures(Systematic review on the comparison of complications and reoperations after plaster casting, K-wire fixation, Kapandji intrafocal pinning, external fixation, dorsal plating, volar plating and intramedullary fixation of extra-articular distal radius fractures in adults).

    I collected all the data and got stuck on the statistics. I tried to look for it in books, Google and YouTube but couldn't find the necessary information.

    How do I pool the data shown below? And secondly how do I compare the pooled data from different treatment modalities? Your help would be highly appreciated!!!!

    Study------ EF complications ---EF total study population--Compl Rate
    1------------3-----------------------23-----------------------------0,130434783
    2------------2-----------------------20-----------------------------0,1
    3------------ 8-----------------------30-----------------------------0,266666667
    4------------ 1-----------------------16-----------------------------0,0625
    5------------ 31---------------------39-----------------------------0,794871795
    6------------ 26---------------------75-----------------------------0,346666667
    7------------ 22---------------------75-----------------------------0,293333333
    8------------ 0-----------------------20 -----------------------------0

    Total-------93----------------------298---------------------------0,312080537
    http://community.cochrane.org/tools/...man-5-download

    This (RevMan) is a pretty popular program for helping organize and analyze a systematic review with meta analysis. If you read it carefully (and use reference materials available with it), it's been said to be fool proof (at least in terms of helping newer researchers format and organize the review). There is a section for outcomes and the relevant data. You just need to select the output (Random-Effects Odds Ratios, for example). Give this a shot and take time to familiarize yourself with it's supplemental material.

  3. The Following User Says Thank You to ondansetron For This Useful Post:

    Rotterdam (01-27-2017)

  4. #3
    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: Pool data and compare data for my systematic review

    Sounds like a lot of treatment comparisons. How many in particular? Also, if you are comparing more than 2 treatments you are supposed to correct your alpha level to address false discovery.


    People also like the Borenstein book which a free PDF is available online.
    Stop cowardice, ban guns!

  5. The Following User Says Thank You to hlsmith For This Useful Post:

    Rotterdam (01-27-2017)

  6. #4
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    I have data on 5 treatment modalities (external fixation, volar plate osteosynthesis, k-wire fixation, intramedullary nail fixation and plaster cast fixation).

  7. #5
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    Thank you for your helpful answer. Hopefully it will be idiot proof enough for me!

  8. #6
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Pool data and compare data for my systematic review

    Quote Originally Posted by Rotterdam View Post
    Thank you for your helpful answer. Hopefully it will be idiot proof enough for me!
    When you spend some time in the software, it'll start to click. It pretty much resembles the format of a manuscript that you can click through to add and organize your information. Good luck!

  9. The Following User Says Thank You to ondansetron For This Useful Post:

    Rotterdam (02-01-2017)

  10. #7
    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: Pool data and compare data for my systematic review

    Do you plan to compare all treatment modalities to one of the treatments or all combinations. This is something you are supposed to decide a priori (before starting), before seeing the results. If you are looking at all combinations, that would be like 10 comparisons, meaning if you used a bonferroni correction a traditional 0.05 cut-off would be changed to a 0.05/10 = 0.005 cutoff level. Also, you will need to think about how many studies may be available for each treatment modalitie. In that there needs to be enough direct comparisons between modalities to power the comparisons. And given in this last sentence is that there needs to be direct comparison of treatment, unless you want to do network meta-analyses which are still developmental when it comes to SE, and would also need to be well powered.


    I guess my take home message is to write out your objectives and planned protocol before running tests, to ensure you aren't biasing the results.
    Stop cowardice, ban guns!

  11. The Following 2 Users Say Thank You to hlsmith For This Useful Post:

    ondansetron (01-27-2017), Rotterdam (02-01-2017)

  12. #8
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    We would like to compare all combinations, because we want to know (looking at the complication rate and reoperation rate) what the best treatment modality is. In the external fixation group we have 93 complications and 298 patients, in the intramedullary nail 43 complications and 203 patients, plaster cast 77 complications and 372 patients, k-wire fixation 73 complications and 285 patients and volar plate 32 complications and 143 patients. I am currently trying to use RevMan, but this is quite difficult because I want to compare all combinations. What is the thing I can do at this moment?

  13. #9
    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: Pool data and compare data for my systematic review

    So once again you need to make sure you correct your alpha, so instead of odds ratios and 95% confidence intervals, you would have 99.95% confidence intervals. This would mean your confidence intervals should be even wider than normal.


    Also, I am not well versed in meta-analysis, but I know that with pooling/random effects, it is not just the sample sizes but also the number of studies. If you do not have many studies you have lower confidence, meaning larger standard errors. This can lead to instability in estimates and even wider confidence intervals I believe.
    Stop cowardice, ban guns!

  14. #10
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    Thank you all very much for your great help. I tried pooling and comparing the pooled data with RevMan, Rstudio and SPSS (with Wilson Macros) but still was not able. RevMan was very easy to use, but I could not compare Plaster Cast with Volar Plate for example, because the data for Plaster Cast was from different studies than the data for Volar Plating. It only worked if the included studies all compared Plaster Cast with Volar Plating directly. I tried using R with metafor (Meta-Analysis Package for R) but when using this program for the first time is very hard and similar to SPSS with Wilson Macros. Can someone PLEASE help me again. I know I am asking a lot of you guys, but I tried finding answers last week, the weekend and pulled an all nighter but still was not able to.

  15. #11
    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: Pool data and compare data for my systematic review

    Say you have three interventions: A, B, C. And A has been compared to B and B compared to C, but A and C have never been compared.


    Are you trying to compare A and C for some studies? If So, that is called a Network Meta-Analysis and you can't use traditional approaches.
    Stop cowardice, ban guns!

  16. The Following User Says Thank You to hlsmith For This Useful Post:

    Rotterdam (02-01-2017)

  17. #12
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    If I try to compare all treatments with Plaster Cast treatment (the golden standard), four comparisons, A with B, A with C, A with D and A with E all with the fixed effects model. How do I pool the data of each treatment and with which test do I compare the treatments with a cutoff level of 0.05/4=0.0125

  18. #13
    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: Pool data and compare data for my systematic review

    And A has been compared to all of the others in the studies, yes?


    If yes, you can just do 4 meta-analyses. You can test for heterogeneity, I^2, which dictates the need for random effects. Have you done this?
    Stop cowardice, ban guns!

  19. #14
    Points: 449, Level: 8
    Level completed: 98%, Points required for next Level: 1

    Posts
    19
    Thanks
    5
    Thanked 0 Times in 0 Posts

    Re: Pool data and compare data for my systematic review

    The problem is I do not know how I can pool all the complication rates of A.

  20. #15
    Points: 1,741, Level: 24
    Level completed: 41%, Points required for next Level: 59

    Posts
    230
    Thanks
    37
    Thanked 68 Times in 59 Posts

    Re: Pool data and compare data for my systematic review


    Quote Originally Posted by Rotterdam View Post
    The problem is I do not know how I can pool all the complication rates of A.
    Are you saying "A" is one outcome, and you're trying to combine the information on the different rates of "A" reported in your included studies? If so, RevMan can do this for you. You create an outcome, and then you add studies to that outcome. Once you've done that, you can enter the raw data into the appropriate row relating to each study. Then, you'll see the pooled estimate (should be using relative risk if you're pooling prospective studies [prospective cohort or RCT, for example]). I've also heard that using a random effects assumption is typically more reasonable than a fixed effects for the pooling.

    Does this help?

+ Reply to Thread
Page 1 of 2 1 2 LastLast

           




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