random intercept coefficient interpratation GLIMMIX

PLB — Fri, 22 May 2020 18:58:49 GMT

Hello all,

I'm trying to assess interhospital variations in hospital admission (binary variable). We suspect a cluster effect so I was a mixed logistic model with hospital center as random intercepts. Using a solution statement I was able to obain coefficient estimate for each center but I'm not sure how to interpret these.

In a mixed linear model these would be the difference with he mean 'hospital' fixed intercept I think, but not sure for logisitc equivalent.

Thanks

Re: random intercept coefficient interpratation GLIMMIX

StatsMan — Sun, 24 May 2020 17:13:04 GMT

The interpretation is the same as for the linear mixed model. The SOLUTION option on the RANDOM statement gives you the random deviations for each level of the SUBJECT= effect from the overall model intercept.

Posting code in the future will help you get a quicker answer 🙂

topic random intercept coefficient interpratation GLIMMIX in Statistical Procedures

random intercept coefficient interpratation GLIMMIX

Re: random intercept coefficient interpratation GLIMMIX