Is test of proportion the correct option

[daniela Albers]

Hello everyone, I need some help. I have a dataset with 36 patienes, 18 received treatment A and 18 treatment B. I ask if they had pain (0=no pain, 1=pain) at 3 times (initial, 6 month and12 month). And I want to compare the answer for each group (that is if there is a change for pain response in time for each treatment) but also compare between treatment. I was thinking in doing a Test of Proportions to compare between groups, for example to compare answers at 6 months between treatment A and B. And do a Test for trend across ordered groups (ntrend) to see if there is difference in pain response across time (for each group separaterly). But to be honest I am not quite sure if I am doing the write thing?

I have attached some examples to help understand my doubt.
Thanks all for the help,
Daniela

[Carlo Lazzaro]

Daniela:
you could also consider -xtlogit- with three-way interaction:

xtlogit pain i.group##i.treatment##i.times