I am working with the national inpatient sample database. It is a complex survey design with one stage power sampling unit - hospitals - among stratas, with national estimates weights on hospitals.

I want to compare the prevalence of a condition in a case cohort that is 0.5% of that sample carrying another baseline condition, essentially giving me a control cohort consisting of millions observations vs a few thousands from the case cohort. What are the implications of this?

What is the difference if i subset the sample first to the condition in question and then subdivide in groups with and without my baseline condition? How would that influence my odd ratio calculation?

Thank you.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.