Do you have an indexing variable for the repeated measures on the patient?  Also, you have ''related" as the response variable, but included in the CLASS statement.  I would recommend removing it from the CLASS statement I have a couple of other changes that are recommended for convergence problems (restatements of the RANDOM statement), and I recommend using method=laplace. If so, a crude approach might be the following: proc glimmix data=events method=laplace;      class patient passes physician priors site visitindex;      model related = priors passes postwt bmi/dist=binary link=logit ddfm=bw solution;      random visitindex/subject=patient type=cs solution; /*patient is repeated measure*/      random intercept/subject=physician solution; /*physician and site are random effects*/      random intercept/subject=site solution; run; Other questions: Are physicians located within site, or might a given physician be at one or more sites? How many patients are included?  If this number is substantially greater than the number of physicians and sites, then the G-side implementation above (conditional approach) could be replaced with an R-side parameterization (marginal approach), such as: proc glimmix data=events;      class patient passes physician priors site siteindex;      model related = priors passes postwt bmi/dist=binary link=logit ddfm=bw solution;      random siteindex/residual subject=patient type=cs solution; /*patient is repeated measure*/      random intercept/subject=physician solution; /*physician and site are random effects*/       random intercept/subject=site solution; run; Good luck. Steve Denham ... View more