We propose a method for estimating parameters in generalized linear models
when the outcome variable is missing for some subjects and the missing data
mechanism is non-ignorable. We assume throughout that the covariates are f
ully observed. One possible method for estimating the parameters is maximum
likelihood with a non-ignorable missing data model. However, caution must
be used when fitting non-ignorable missing data models because certain para
meters may be inestimable for some models. Instead of fitting a non-ignorab
le model, we propose the use of auxiliary information in a likelihood appro
ach to reduce the bias, without having to specify a non-ignorable model. Th
e method is applied to a mental health study.