Meta-analysing differences between sensitivities and specificities

[Jawdat Abdulla]

I appreciate any answer from STATA meta-analysis experts

I have a data set of 4 diagnostic accuracy studies each with 2 subgroups stratified according to body mass index. My question is how to meta-analyse the differences between sensitivities and specificities of the two subgroups for statistic significance ?

Would it be correct to use:

admetan tp1 (tp1+fn1) tp2 (tp2+fn2), OR random for sensitivities difference, and:

admetan tn1 (tn1+fp)1 tn2 (tn2+fp2), OR random for speficities difference ?

Or any other suggestion ?

Thanks
Jawdat Abdulla
Copenhagen
Denmark

[Tiago Pereira]

Hi, Jawdat.

I believe this question should be asked in the general forum (it does not related to mata).

You can use -meqrlogit-


Check this material:

https://methods.cochrane.org/sdt/sit...Feb%202023.zip

Cheers!

T

[Jawdat Abdulla]

Hi
Thanks too much
BR
Jawdat