[Mah Tab]

Hi David

I am using metan to do a meta-analysis and create forest plots for it. I have data on hazard ratios with their CI. In the output from Stata, I can see that tau-squared comes with confidence interval. I am able to add the tau- squared and I-squared to the forest plot. However, I don't know how I can incorporate the confidence intervals in the forest plot. What I want is that when I create the forest plot, the heterogeneity metrics along with their confidence intervals be reported in the plot.
This is the code I used.

metan logHR loglowerCI logupperCI, eform effect(Hazard ratio) random(ebayes) forestplot(hetinfo (isq tausq) extraline(yes))

Best regards,
Mah

[spencer keene]

Hello All,

I have an issue where I would like to fill the boxes with colours according to a variable called exp_comp which has two subgroups. The final forest plot should have two estimates (one red and one blue) per group. admetan achieved something similar with the addition of plotid(exp_comp) forestplot(ci1opt(lcolor(red)) box1opt(mcolor(red)) ci2opt(lcolor(blue)) box2opt(mcolor(blue)))
However, admetan does not do quite what I would like. My metan command is below:

metan coef se, eform lcols(exp_comp) ///
by(group) nosubgroup rcols(rr_ci p_val2) xlabel(0.75, 1,1.5, 2.0) force astext(72) classic boxsca(50) nostats ///
textsize(120) boxopt() diamopt(lc(black) lw(thin)) pointopt() ciopt(lw(vthin)) /*aspect(1.8)*/ ///
olineopt(lc(black) nobox lw(vthin) lp(shortdash)) nooverall graphregion(fcolor(white))

[spencer keene]

My question above was also posted to StackOverflow: https://stackoverflow.com/questions/...using-stata-16

[Nick Cox]

David Fisher is assiduous about answering on this topic once he knows there's a question. See if that ping works; otherwise email him directly.

Stack Overflow is not a good place for this kind of question, as most people active there on Stata know very little about meta-analysis. I'd delete your SO post.

[spencer keene]

Thanks Nick. I will try to email him tomorrow if no reply here

[spencer keene]

Sorry coming back to my issue. it seems I found the answer here: metan/Examples/Example3.md at main · UCL/metan · GitHub

I will try it now and leave an update