Helllo! I am a research student that is working with survey data to examine the co-occurrence of itch with other co-morbidities. Basically, I am looking at whether or not itch presents with other comorbidities such as PulmonaryCardiac pre- and post-survey. The following code, for example, I think, was used to delineate that:
if(Pre_PulmonaryCardiac = 0 & Post_PulmonaryCardiac = 1) then PulmonaryCardiac = 1;
else if(Pre_PulmonaryCardiac = 1 & Post_PulmonaryCardiac = 0) then PulmonaryCardiac = 2;
else if(Pre_PulmonaryCardiac = 1 & Post_PulmonaryCardiac = 1) then PulmonaryCardiac = 3;
else PulmonaryCardiac = 0;
I'm now trying to create a combined frequency table of results to examine whether itch presented with PulmonaryCardiac regardless of whether its pre- or post-survey. In other words, I want to develop a format that combines any patients that have any instance of a particular variable and set that to 1, if no then 0.
The code I am currently utilizing does not give me the right combined results:
proc freq data = ITCH.CROSS_MERGED1;
where PulmonaryCardiac = 0 | PulmonaryCardiac = 2;
tables rMSA_Q_12 * PulmonaryCardiac / chisq;
format PulmonaryCardiac FORMAT_FOR_COMPOSITES. rMSA_Q_12 rMSA_2bin.;
How should I format a code that would allow me to get frequencies for: (1) no Dx pre and post (2) Dx pre, post, whereever?
Not sure if that's what you want, but a format to divide the data into any/no pulmonary condition can be written like this:
proc format;
value anyPulmonary
0='No pulmonary condition'
1,2,3='Pulmonary condition'
;
run;
You can than apply that format to your PulmonaryCardiac variable.
SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.