Admetan - random or fixed effect?

[Katrine Lawaetz]

Dear StataList,

I am sorry if this is a stupid question.

I have used admetan for my meta-analysis. I have developed it over the years to accommodate the right output.

The code is as follows

admetan lor_cr llci_cr luci_cr if !missing(lci_cr), effect(OR) or by(Subgroup) ///
    forestplot(astext(60) spacing(1.4) nobox lcols(study total) ///
    xlabel(0.2 0.5 1 2 5) cirange(.2 5) range(0.2 5) ///
    favours(Reduced risk of AAA # Increased risk of AAA)) ///
    nowt keeporder     saving(temp)

use temp, clear
label var _LABELS `"{bf:Study & Subgroup}"'
replace _LABELS = `"{bf:"' + _LABELS + `"}"' if _USE==0

label var _EFFECT `"{bf:OR (95% CI)}"'
label var total `"{bf:N}"'
label var _WT `"`"{bf:%}"' `"{bf:Weight}"'"'

forestplot, useopts nostats rcols(_EFFECT)

Now, a reviewer has asked me: The statistical method of creating a 'weighted average' (i.e. meta-analysis) should be described - i.e. 'fixed effects' or 'random effects' methods.

I am not sure - and I have asked both Google and searched the manual and Statalist, what the default is - since I apparently haven't specified it in my code.
Can anyone please help me?

Best regards,
Katrine

[Tiago Pereira]

Katrine, hi.

My admetan is:

. which admetan
/home/tiago/ado/plus/a/admetan.ado
*! version 3.2  David Fisher  28jan2019

If you run your code, you will be able to see the output, which should show the method in bold:

Studies included: 10
Participants included: Unknown

Meta-analysis pooling of aggregate data
using the fixed-effect inverse-variance model

[Katrine Lawaetz]

Thank you, Tiago! I really just had to look closer at the results window... Mine uses the fixed-effect inverse-variance model, so I guess that is the default. Thank you again, K