topic Re: SAS to do Data Analysis Crude OR and Adjusted OR in SAS Programming

SAS to do Data Analysis Crude OR and Adjusted OR

jessica_join — Thu, 29 Mar 2018 22:59:17 GMT

Need crude and adjusted odd ratios with the CIs

I have a project where I have to do a data analysis of a cross sectional study looking at a minimum of 4 confounders and 1 effect modifier on top of the exposure using SAS.

I am supposed to run a proc logistic on this to give crude and adjusted odd ratios . My issue is that the variables all have multiple levels such as my exposure is regions. How do I get the Odd ratio for each individual region? My effect modifier is sex, how do I get the individual OR for male and female?

Please help! I am so bad at SAS and I need to get the data before I can complete the rest of my project.

Re: SAS to do Data Analysis Crude OR and Adjusted OR

Reeza — Fri, 30 Mar 2018 00:08:26 GMT

Use the oddsratio statement. Check out this example - run the code too!

http://documentation.sas.com/?docsetId=statug&docsetTarget=statug_logistic_examples02.htm&docsetVersion=14.3&locale=en

With one note: set param=REF on your CLASS statement.

If if you’re still uncertain post back with specific questions and your code.

Re: SAS to do Data Analysis Crude OR and Adjusted OR

jessica_join — Fri, 30 Mar 2018 15:34:11 GMT

Thank you! But how would I do a stratified analysis to find the individual ORs for each group compared to the outcome?

Re: SAS to do Data Analysis Crude OR and Adjusted OR

Reeza — Sat, 31 Mar 2018 01:39:56 GMT

Do you want stratified logistic regression or are you looking for something like: Men are 1.3X more likely to die before 80 than women?

Stratification is a different process and then you would include that variable in your STRATA statement.

Otherwise, the oddsratio is the number you're looking for here.