Modelling the treated course of schizophrenia: Development of a discrete event simulation model

Abstract

In schizophrenia, modelling techniques may be needed to estimate the long-term costs and effects of new interventions. However, it seems that a simple direct link between symptoms and costs does not exist. Decisions about whether a patient will be hospitalized or admitted to a different healthcare setting are based not only on symptoms but also on social and environmental factors. This paper describes the development of a model to assess the dependencies between a broad range of parameters in the treatment of schizophrenia. In particular, the model attempts to incorporate social and environmental factors into the decision-making process for the prescription of new drugs to patients. The model was used to analyse the potential benefits of improving compliance with medication by 20% in patients in the UK. A discrete event simulation (DES) model was developed, to describe a cohort of schizophrenia patients with multiple psychotic episodes. The model takes into account the patient’s sex, disease severity, potential risk of harm to self and society, and social and environmental factors. Other variables that change over time include the number of psychiatric consultations, the presence of psychotic episodes, symptoms, treatments, compliance, side-effects, the lack of ability to take care of him/herself, care setting and risk of harm. Outcomes are costs, psychotic episodes and symptoms. Univariate and multivariate sensitivity analyses were performed. Direct medical costs were considered (year of costing 2002), applying a 6.0% discount rate for costs and a 1.5% discount rate for outcome. The timeframe of the model is 5 years. When 50% of the decisions about the patient care setting are based on symptoms, a 20% increase in compliance was estimated to save £16 147 and to avoid 0.55 psychotic episodes per patient over 5 years. Sensitivity analysis showed that the costs savings associated with increased compliance are robust over a range of variations in parameters. DES offers a flexible structure for modelling a disease, taking into account how a patient’s history affects the course of the disease over time. This approach is particularly pertinent to schizophrenia, in which treatment decisions are complex. The model shows that better compliance increases the time between relapses, decreases the symptom score, and reduces the requirement for treatment in an intensive patient care setting, leading to cost savings. The extent of the cost savings depends on the relative importance of symptoms and of social and environmental factors in these decisions.