The outcome of psychosis

Schizophrenia spectrum disorders, or psychotic disorders, are considered among the most serious life shortening, burdening mental illnesses worldwide, at a personal, familial and societal level, with approximately a 2- to 3-fold mortality risk as compared to the general population. The symptoms associated with psychosis tend to have devastating effects on global, social, occupational and daily functioning, quality of life and physical health. Illness outcome is an important measure in health care planning. Moreover, patients (and their families) presenting with a psychotic episode at the clinic want to know what to expect regarding the course of the disease. Many factors, including premorbid, clinical, demographic, environmental and genetic characteristics may contribute to the likelihood of a poor outcome after a first psychosis. But to what extent do these factors together explain outcome? At present little is understood of mechanisms leading to an unfavorable but also to a favorable outcome.

In the first part of the thesis, studies are presented examining determinants in childhood and determinants after the development of psychotic illness in relation to clinical outcomes and global functioning. In the second part of the thesis the emphasis is on physical outcomes, in particular metabolic disturbances and smell processing, and its relation to clinical variables, cognition and brain volumes.

By examining precursors that play a role in mental health and physical outcomes of psychotic disorders, it becomes clear that the underlying mechanisms are multifaceted. Taken together, research presented in this thesis highlights that predictors of mental health outcomes and physical outcomes include an interaction of many factors concerning the (intermediate) phenotype such as structural brain integrity (grey matter volumes and white matter pathways wiring), premorbid functioning, demographic characteristics, psychiatric symptomatology, need of care, and (social) cognitive functioning. Some predictor modalities hold special promise, such as reward processing in health outcome prediction; premorbid functioning and (social) cognition in predicting metabolic syndrome and olfactory identification deficit; connectome organization,patients’ needs and present state/lifetime psychotic symptoms in symptomatic and global functioning outcome prediction. Furthermore, it can be concluded that mental health and physical wellbeing are not separable in schizophrenia spectrum disorders. Much of the same predictive factors of mental health outcomes also seem to apply to physical outcomes. The studies presented in this thesis confirm the necessity of a multidisciplinary approach in health care, using mental health and physical outcome as integral treatment target. Future focus should be on personalized outcome prediction. We used machine learning techniques to predict outcome in individual patients with schizophrenia, and although our results are promising, these prediction models need to become more accurate before they can be translated into clinically useful tools. Even more elaborate sets of predictors are necessary, as well as larger multicenter prospective cohort studies for replication, before individualized prediction tools can be developed, tested and implemented in the clinic.