Can I use distrate for clustered data?

[Rimma Myton]

Hello, I have data with number of health conditions entered as binary variables 1 (Yes) or 0 (No), eg. asthma, clustered in sites (site_ID). I would like to use direct age standardisation using age_cat to estimate their prevalence and 95% CI, where the entire cohort is used as standard population (stand_pop). So my code would look something like:

svyset site_ID

svy: tabulate asthma, stdize(age_cat) stdweight(stand_pop_weight) ci

My understanding is that Stata uses 'standard' CI calculation assuming normal data. I would like to have CI based on Dobson method as more accurate. I know that distrate would provide such results. However, I am not sure if I can use it or adopt it for clustered data. Could you please share your thoughts on whether I should use

distrate

or just use the

svy

results or manually calculate Dobson CI? Any feedback is much appreciated.

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - Stata code in code delimiters, readable Stata output, and sample data using dataex.

Help with a user-written procedure depends on someone active on the list using that procedure. It appears that no one is available. You might try the procedure's author.

Sometimes, it is helpful to run an estimate using a data set from a text so you know what you should be getting and see if your estimation approach gives you the right answer.