Packages for making odds ratio plot

[Sonnen Blume]

Hi, I just came across an article that used Stata for all analyes and reported the following odds ratio chart. Can anyone guess which package was used to create this. I couldn't find this anywhere in the diaglog boxes.
Thank you.

[Weiwen Ng]

I would be interested in the question as well. In meta-analysis, the graph would probably be called a forest plot. I have heard the name caterpillar plot, but I believe that's often used to visually show the distribution of random effects after fitting a random effects model. There is a https://www.stata.com/manuals/metametaforestplot.pdfhttps://www.stata.com/manuals/metametaforestplot.pdf as part of the meta analysis set of commands in Stata 16, but I can't immediately see how it would be adapted to this use case.

[Alexander Rodriguez]

Dear Sonnen Blume,
I too would be interested as to what you have learned. Stata has many great forest plot commands but not that many for when trying to use one outside the context of a meta-analysis.

I have found margins and marginsplot difficult to manipulate if you are not conducting regression analysis (for instance, I want to plot crude IRR using iri).

Good luck!

[William Lisowski]

This topic suggests a way of using meta-analysis tools without actually doing a meta-analysis.

https://www.statalist.org/forums/for...tic-regression

[David Fisher]

Hi Sonnen,

My command forestplot (available as part of the admetan and ipdmetan packages, from SSC) was designed specifically to address the problem of forest plot -style graphs outside the context of a meta-analysis.

I hope you find it useful! Please let me know if you have any questions about it.

Thanks,

David.

[Sonnen Blume]

Hi Sonnen,

My command forestplot (available as part of the admetan and ipdmetan packages, from SSC) was designed specifically to address the problem of forest plot -style graphs outside the context of a meta-analysis.

I hope you find it useful! Please let me know if you have any questions about it.

Thanks,

David.

Hi David,
Thanks indeed for your message. I'm sorry for my late response. I actually used the

metan

and

coefplot

packages before to make odds ratio plots. Both of the processes feel like a bit of detour, especially

metan

and the like packages that require manually organising the ORs, LCIs and UCIs to produce the graph. I'd feel encouraged doing so if it could produce a graph as informative & handsome as the one attached at #1. I contacted the authors, but no luck there, and here, so far. So I'll have to continue the search.

Best.

[David Fisher]

Hi Sonnen,

Did you take a look at forestplot ?

Also, you say the "...processes feel like a bit of a detour... require manually organising the ORs....". Could you tell me more about the specific structure of your data, so that I can comment more on this? Hopefully we can find the right way forward for you.

Many thanks,

David.