Hi,

I have some health centre data from Africa where each record consists of fields that represent monthly totals of important events, e.g., the number of deliveries that month in that health centre, plus the number of caesarian sections etc. Here's an example of a few records each containing a few variables.
health centre year month No. of referrals No. of c-sections Number of mothers with preeclampsia
d 2016 7 0 11 0
d 2016 8 0 20 0
g 2016 7 15 5 0
g 2016 8 34 0 0
k 2016 7 0 0 3
k 2016 8 3 0 0
m 2016 7 4 0 0
m 2016 8 6 0 0
n 2016 7 7 0 0
n 2016 8 1 0 0

How do I analyse such data? for example, health centre 'd' had 11 c-sections in July and 20 in August. Looking at the data by time variables, there were (11+5) 16 C-sections in July and 20 in August. What Stata commands can do a sum of such counts, especially by important stratification variables like year or health centre?

This is not a straightforward wide to long dataset transformation because each variable has a different meaning. It's essentially a long dataset where each record has counts rather than individual data.

I'm stumped!

Thanks for any help.

Regards, John LeBlanc
__________________________________________________ _______

John C. LeBlanc, MD, MSc, FRCPC
Associate Professor
Pediatrics, Community Health and Epidemiology, Psychiatry
Dalhousie University

IWK Health Centre Work phone: 902 470-8930
5850 University Avenue Work fax: 888 557-0867
Halifax, Nova Scotia Email: John.LeBlanc@Dal.Ca
B3K 6R8 CANADA Pager: 902 470-8888
__________________________________________________ _______