topic Re: Getting prevalence ratio with categorical survey data? in Statistical Procedures

Getting prevalence ratio with categorical survey data?

SAS93 — Thu, 08 Apr 2021 11:29:20 GMT

I'm using complex survey data to look at associations between a binary outcome and dichotomous "exposure" variable (male/female) that's also controlling for several other sociodemographic variables. This analysis is based on/replicating Ekenga 2019 (https://doi.org/10.1007/s10900-019-00638-5)

I'm planning to also use proc surveylogistic (so...logistic regression), to build logistic models, as was done in the original paper. But I need to find the prevalence ratio/relative risk, not the odds ratio. How do I do this? What would the coding look like?

Re: Getting prevalence ratio with categorical survey data?

StatDave — Thu, 08 Apr 2021 14:45:47 GMT

You should be able to use the NLMeans macro in the same way as shown in this note, just using PROC SURVEYLOGISTIC rather than PROC LOGISTIC.

Re: Getting prevalence ratio with categorical survey data?

maciulei — Tue, 12 Sep 2023 19:41:00 GMT

Hi, would you happen to have the NLMeans macro? Thanks!

Re: Getting prevalence ratio with categorical survey data?

StatDave — Tue, 12 Sep 2023 20:24:52 GMT

The NLMeans macro is shipped with recent SAS releases, so you should have a version of it available. However, you can always download the latest version from the NLMeans documentation.