Abstract

In mental health care clinical pathways determining the form of therapy that has to be administered are highly controversial and have not been officially established. However, surveys indicate that different treatments of the same mental disease can lead to different therapeutic outcomes. This has a direct influence on the number of future patients of the health care providers and thus on the system's input and output including treatment costs and quality of care.
In this paper we use computer simulation to examine the outcome of different treatments for mental disorders in a psychiatric hospital. The compliance of the patient, that is the motivation for taking part in the treatment, is crucial for the success of a therapy. In case of a discontinuation the probability of the mental disturbances occurring again is higher than in the case of a successful completion. As the in-patients' parameter compliance is changing continuously during the treatment we need a simulation technique allowing continuous attributes of entities in a discrete-event-driven system. We use the introduced concept of hybrid tokens in stochastic Petri nets for modelling the treated in-patients with both discretely and continuously changing attributes. For that purpose, a sufficient mathematical description of the in-patients' mental parameters and their development over time has to be derived. In doing so, a comprehensive statistical analysis has to be performed. The provision system and the resulting Petri net are comparatively simple. Instead, a high number of tokens that has to be created and the computation of their attributes during the simulation run are the challenges that we are facing. After building the simulation model of the therapy processes we were able to run experiments by variegating characteristics and treatments of the patients and observe the system's output.
We believe that the implemented model can provide a decision support for physicians and therapists. It will enable the estimation of the therapeutic outcome and thus the choice of the most promising treatment accord-ing to the patient's mental disturbances.