Relative Risk Plot in Stata

[Melissa McCarthy]

Can someone tell me how to plot relative risk and 95% CI in stata. I used the cs command to estimate these parameters but am having trouble plotting them. Thank you!

[Dave Airey]

You can combine plot types in Stata to get the effect you want, but better would be to first search Statalist:

http://www.stata.com/statalist/archi.../msg00421.html

[Bruce Weaver]

Hello Melissa, and welcome to the forum. Here are a couple of possibilities that come to mind. (Bear in mind that I'm a relative newbie to Stata, and some of the grizzled old veterans may come along with better suggestions after they read this!)

* Use example from this UCLA page:
* https://stats.idre.ucla.edu/stata/faq/how-can-i-estimate-relative-risk-using-glm-for-common-outcomes-in-cohort-studies/
clear
use https://stats.idre.ucla.edu/stat/stata/faq/eyestudy
cs lenses carrot
* return list
generate Upper = r(ub_rr) // 95% CI upper bound
generate Lower = r(lb_rr) // 95% CI lower bound
generate RR = r(rr) // risk ratio (RR)
generate byte Exposed = 1 // variable to use on X-axis of plot
twoway rcap Upper Lower Exposed || scatter RR Exposed

* Use -glm- to estimate RR, then use -coefplot- to plot it
glm lenses i.carrot, fam(bin) link(log) nolog
*coefplot, drop(_cons) yline(0) ytitle(ln(Risk Ratio)) vertical
coefplot, drop(_cons) yline(1) ytitle(Risk Ratio) vertical eform

Here are some links that may help you decipher what is going on in those examples.

• https://stats.idre.ucla.edu/stata/fa...ohort-studies/
• http://www.philender.com/courses/cat...es3/rrisk.html
• http://www.stata.com/meeting/germany.../de14_jann.pdf

HTH.

[Andrew Lover]

Try -eclplot- (downloadable from SSC); and if you want to use the dialog box after downloading it type -db eclplot-.

Great tutorial here: http://www.imperial.ac.uk/nhli/r.new...ewson_ohp1.pdf

[Tiago Pereira]

You can also check out -metan-, specially if you have subgroups. It plots RR and 95% CI, and is very flexible.

http://www.stata-journal.com/sjpdf.h...clenum=sbe24_2

Tiago

[David Fisher]

...or, alternatively, -admetan-, which is effectively an update of -metan- (which has not been updated for 7 years or so, as far as I'm aware!). Note that -admetan- is part of the -ipdmetan- package, downloadable from SSC.

See: http://www.statalist.org/forums/foru...13#post1392313

David.

[Elham Mohebbi]

Hello,
could anyone tell me how to use "ipdover" for logistic regression? Is it possible to use ipdover for case-control studies?