Modelling patient flow in health care systems is considered to be vital in understanding the operational and clinical functions of the system and may therefore prove to be useful in improving the functionality of the health care system, and most importantly provide an evidence based approach for decision making, particularly in the management of the system. In this paper, we introduce a nonproportional cumulative odds random effects model for patient pathways by violating the proportional assumption of the cumulative odds model. Using the probability integral transform, we have extended this to cases where the random effects are not normal, specifically gamma and exponentially distributions. Some of the advantages of this is that these models depict changes in wellbeing (frailties) of patients as they move from one stage of care to the other in time. This is an hybrid extension of our earlier work by jointly including pathways and covariates to explain probability of transition and discharge, which could easily be used to predict the outcome of the treatment. The models here show that the inclusion of pathways render patients characteristics as insignificant. Thus, pathways provide a source of useful information about transition and discharge than patient characteristics, especially when the model is applied to a London University Neonatal Unit dataset. Bootstrapping was then used to investigate the stability, consistency and generalizability of estimated parameters from the models.