Identifying twins that have a feature in common

[David Metcalfe]

Hi all,

I am working with a dataset that contains information on a number of twins. At least one of the twins in each pair has disease X and I am trying to work out whether this disease has a strong genetic component. One thing that I need to do is create a variable that identifies concordant (i.e. those that both have the disease) and discordant (i.e. those in which only one twin has the disease) pairs.

Each individual comes with a unique number to identify their twin pair, a number to identify them within the pair (1 or 2), and a variable indicating whether or not they have the disease (0 or 1).

Can anyone suggest a way in which I could create a variable that highlights concordant and discordant twins?

Any help would be very welcome!

David

[Devra Golbe]

David,

I think something like the following untested code should work:

bysort twin_pair_id: egen maxX=max(hasX)
bysort twin_pair_id: egen minX=min(hasX)
gen cordant=0
replace cordant=1 if maxX==minX

Best,
Devra

[Nick Cox]

Consider

-bysort twinid : gen both = x[1] == 1 & x[2] == 1

Also search for Tip 71 on dyads in the Stata Journal