Meta-analysis: counting the number of patients of each group.

[Bruna Holanda]

Hi, I am struggling in the beginning of my thesis.

I have a metanalysis:
I have 20 studies from low and middle-income countries (LMIC)
and 88 studies from High-income countries (HIC)

Those studies from LMIC have in the total 1721 patients and from HIC I have 4641 patients.

1. How I am supposed to count and do the analysis based on the number of patients and don't on the number of studies?
I have one variable patients and one variable income; but how I connect both and make the analysis direct from the patients of each income?

2. When I do a tabstat, is that possible to see the percentage of this number related to the total?
For example: 1580 is equal to how many % of 1721 below? is that possible to have it directly or should I do it manually?

Thank you so much,
I appreciate your help.

Best,
Bruna

[George Ford]

I don't know what you're asking in 1.

For 2, you probably should learn the -table- command.

A kluge on tabstat would be (maybe):

tabstat ....., statistics(sum) save
matrix R1 = r(Stat1)
matrix R2 = r(Stat2)

forv i = 2/8 {
    matrix R1[1,`i'] = R1[1,`i']/R1[1,1]
    matrix R2[1,`i'] = R2[1,`i']/R2[1,1]
}
matrix list R1
matrix list R2