Must SUDAAN be used for Survival Analysis of NHANES data?

[closetcoer]

I've inherited a decade-old SAS program that analyses NHANES survey data. The creator of the program used both Cox Regression for survival analysis (with which I'm familiar) and SUDAAN (with which I'm not familiar).

My question very basic conceptual question (sorry if it's too basic) is whether it is necessary to use SUDAAN to perform survival analysis with survey data like NHANES, given the availability of "proc surveyphreg" and "procesurveylogistic" in SAS 9.4?

Thanks!

[Reeza]

It depends on the type of 'survival analysis'. SAS can do more than it did a decade ago. Can you run both methods and see if you get the same/similar results and then make the decision from there going forward?

[closetcoer]

Hi Reeza:

Thanks for your reply - it sounds like the newer SAS procs are sufficient as a start.

I was hoping to avoid running both as I don't have experience with SUDAAN but am thinking I might have to learn how to use it. 

Any recommendations for a good primer? 

Thanks.

[closetcoer]

Oops, also meant to specify that I am particularly focused on Cox Proportional Hazards regression.

[Reeza · Accepted solution]

I would recommend finding a SUDAAN forum, but Google may. I did find this that may be of interest, if you didn't already find it: https://www.lexjansen.com/sesug/2015/42_Final_PDF.pdf
Its a comparison of PROC SURVEYPHREG and SUDAAN for survival.

[closetcoer]

Thanks (again)! Will check it out.

[FreelanceReinh]

Hi @closetcoer,

Glad to see that Reeza answered your questions. Then it would help later readers if you marked her most helpful reply as the accepted solution, not your own "thank you" post. Could you please change that? It's very easy: see instructions here.

[closetcoer]

Thanks for the feedback - I think I fixed it.