Dear Allstat users,
I wondered if some body could help with my problem of modeling:
We are developing a risk model for infection after surgery using logistic regression in the frame work of generalized linear model techniques. The problem we face is, for some of the patients the surgery is repeated (but not for all), which add up an extra variability to the data. I was thinking to add the repeated surgery as a co-variate but we think that the unit of measurement, for the patients with repeated surgery, is the surgery per patient rather than patient. Is there any way we can handle this problem in generalized linear model for logistic regression.
GEE can be used for repeated measure, but the problem is the surgery is repeated for some of the patients but not for all, so I don't think GEE or mixed effect model will work properly.
I will apreciate any help,
thank you in advance,
Muhammad Riaz
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|