Meta-analysis of diagnostic test accuracy studies using midas

[Brandon Galm]

Hello everyone,

I have tried searching for this but I cannot seem to find an answer.

I am doing a meta-analysis of diagnostic test accuracy studies. I am including several different tests, and for each test, there are numerous studies (sometimes >50). Thus, the forest plot can become quite messy.

I am interested in ONLY showing the summary/combined graph (see image, the part in the red square). Then, in my manuscript, I will show only the summary graph for each test. The individual, large forest plots will be available as a supplement.

Can someone please help me out with this? Thank you so much in advance!

Command:
midas tp fp fn tn, id(Study2) ms(0.75) ford bfor(dss)

[Marcos Almeida]

I think a graph is mainly due to convey "lots of information" within a picture. For the overall measure, just informing literally the sensitivity as well as the specificity (with 95% CIs) would wisely provide less space than putting this information in a graph. I know it is up to you to decide about this, and you may just type the numbers in a new dataset and, lo and behold, create a plot with the coefficients. But please beware that, shall the manuscript be sent to a journal, some reviewers would promptly protest against this strategy. And the reason gets crystal clear just by taking a look at the whole graph shared in #1.

[Brandon Galm]

Thank you very much for your response.

The issue is this: it's just way too crowded. I have another figure that has even more studies.