Missing data due to patient dropout is a common problem in clinical trials. While likelihood-based methods offer protection against biased inference in cases where dropout is ignorable, they do not protect against bias in the presence of non-ignorable dropout. In this talk, we present methods that accommodate non-ignorable dropout using readily available software. The basic approach entails jointly modeling the outcome variable and time to dropout using a shared parameter (SP) model. The SP model will be illustrated using data from the MDRD study, a RCT of the effects of diet and blood pressure control on progression of renal disease. We then apply a SP model to the analysis of ordinal data from the NIMH Schizophrenia Collaborative Study in which we compare treatment related changes in disease severity over time with adjustment for possible non-ignorable patient dropout.