Hello,
I have undertook a meta-analysis in Revman 5.3 for the effect of different pollutants on asthma. I am wondering if it is possible to show the different comparisons on one plot rather than separate ones?
Can anyone please advise?
Many thanks
Haneen.
To my knowledge, there is not a way to do this in Revman. I am assuming you are looking at sub-group analysis? IF this is the case, you can generate all of the sub-groups together in one plot.
Are you talking about multiple treatment comparison meta-analysis? Unfortunately, RevMan cannot do that.
I think the answer is YES.
Assume your have input all the results (e.g. a binary outcome of having asthma or not) from different pollutants in RevMan.
1. In the left tree panel, right-click the "Data and analyses", choose "Add comparison" and name it "ASTHMA" for example.
2. Then "Next", choose "Add an outcome under the new comparison", then "Continue".
3. Since we are binary outcome, choose "Dichotomous", and also name it and define some details etc. For example, naming as "Sub-group analysis".
4. Right-click the newly created sub-tree "Sub-group analysis", you could find and click "Add subgroup". Then you can add different sub-group by pollutants e.g. CO2, PM2.5, NO2 etc. one by one. Be careful, only by right-click the "Sub-group analysis" you will find the "Add subgroup" option.
5. Finally, right-click the newly created sub-group, for example "PM2.5", you can "Add study data" under them. Then the final forest plot will be a sub-group style with separate pooled result for each sub-group.
6. Very important one. Remember right-click "Sub-group analysis", go the "Properties"-"Analysis details"-"Total", choose "Subtotals only". It is meaningless if we report a total pooled result including different pollutants.
Pls feel free to ask if I have not written clear enough. Cheers! : )
Thank you all - Charles this has worked this is exactly what I wanted - thank you very much for your help
Best wishes,
Haneen.
Interesting to see RevMan can do this...I was content using Photoshop or Paint to organise everything into one plot!
Rocco
Hello
Is there any way to delete the numbers at the beginning of each subgroup name? Numbers do not match the actual order in the paper.
Thanks
How many studies are needed to perform subgroup analysis? is there any restrictions with this regards?
Although this is an old post, thanks Charles, this has really helped me as i was stuck.
Old post but I have the same issue than Jordan Carolina Villa.
Does anyone get the answer?
Is there any way to delete the numbers at the beginning of each subgroup name? Numbers do not match the actual order in the paper
Thank you!
I am sorry can anyone explian me how to merge the forest plots i do not understand
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis