IF you made absolutely no changes to the SAS program then talk to your SAS admin about what changes may have been made in your SAS environment.     If you have made any changes to the ODS statements in the program this would be, from the very limited code and no log, my main concern: Make sure that there is at least one ODS destination active at the time the Univariate code attempts to execute (you show code closing one, if there are no others such as ODS HTML then there is not ODS output to generate). Also, instead of ODS EXCLUDE ALL I might suggest ODS SELECT only the output you want instead.   My suspicion is that somewhere later in the code you do open another ODS destination other than Listing, or reset one when restarting, that is open again.   Otherwise I would comment out that shown section of the code and run the program and look very closely at the log for the entire job for any warnings or errors that may occur prior to this step. ... View more

Perhaps the easiest would be to consider the demographic variables as fixed covariates.  Include them in a CLASS statement and in the MODEL statement.   If you wish to treat the demographic variables as random effects, you will probably have to try coding in NLMIXED to handle both the distribution and censoring issues. GENMOD and GLIMMIX do not correctly address censoring.   SteveDenham ... View more

Hi Rick,   I am familiar with how to manipulate formats in order to get the desired reference level. However, I am having another issue. I have a categorical variable with 16 levels and every time I run my model (lifereg), regardless of how the values are formatted, it outputs 4 of the levels as reference groups. I only want one level to be a reference group.   Any advice would be appreciated.   Best, Sara ... View more