Recently, there has been increased focus on sub-threshold stages of mental disorders, with attempts to model and predict progression to full-threshold disorder. Given this considerable research attention and clinical significance, it is timely to analyse the assumptions of theoretical models in the field. Research into predicting onset of mental disorder has shown an overreliance on one-off sampling of cross-sectional data (i.e., a "snapshot" of clinical state and other risk markers) and may benefit from taking dynamic changes into account. Cross-disciplinary approaches to complex system structures and changes, such as dynamical systems theory, network theory, instability mechanisms, chaos theory and catastrophe theory, offer potent models that can be applied to emergence (or decline) of psychopathology, including psychosis prediction and transdiagnostic symptom emergence. Staging provides a useful framework to research dynamic prediction in psychiatry. Psychiatric research may benefit from approaching psychopathology as a system rather than a category, identifying dynamics of system change (e.g., abrupt/gradual psychosis onset), factors to which these systems are most sensitive (e.g., interpersonal dynamics, neurochemical change), and individual variability in system architecture and change. The next generation of prediction studies may more accurately model the highly dynamic nature of psychopathology and system change, with treatment implications, such as introducing a means of identifying critical risk periods for mental state deterioration.

For over a decade a transdiagnostic clinical staging framework for youth with anxiety, mood and psychotic disorders (linked with measurement of multidimensional outcomes), has been utilised in over 8,000 young people presenting to the enhanced primary (headspace) and secondary care clinics of the Brain and Mind Centre of the University of Sydney. This framework has been evaluated alongside a broad range of other clinical, neurobiological, neuropsychological, brain imaging, circadian, metabolic, longitudinal cohort and controlled intervention studies. This has led to specific tests of its concurrent, discriminant and predictive validity. These extensive data provide strong preliminary evidence that: i) varying stages of illness are associated with predicted differences in a range of independent and objectively measured neuropsychological and other biomarkers (both cross-sectionally and longitudinally); and, ii) that earlier stages of illness progress at variable rates to later and more severe or persistent disorders. Importantly, approximately 15-20% of those young people classed as stage 1b or ‘attenuated’ syndromes at presentation progress to more severe or persistent disorders. Consequently, this cohort should be the focus of active secondary prevention trials. In clinical practice, we are moving to combine the staging framework with likely pathophysiological paths (e.g. neurodevelopmental-psychotic, anxiety-depression, circadian-bipolar) to underpin enhanced treatment selection.

For over a decade a transdiagnostic clinical staging framework for youth with anxiety, mood and psychotic disorders (linked with measurement of multidimensional outcomes), has been utilised in over 8,000 young people presenting to the enhanced primary (headspace) and secondary care clinics of the Brain and Mind Centre of the University of Sydney. This framework has been evaluated alongside a broad range of other clinical, neurobiological, neuropsychological, brain imaging, circadian, metabolic, longitudinal cohort and controlled intervention studies. This has led to specific tests of its concurrent, discriminant and predictive validity. These extensive data provide strong preliminary evidence that: i) varying stages of illness are associated with predicted differences in a range of independent and objectively measured neuropsychological and other biomarkers (both cross-sectionally and longitudinally); and, ii) that earlier stages of illness progress at variable rates to later and more severe or persistent disorders. Importantly, approximately 15-20% of those young people classed as stage 1b or ‘attenuated’ syndromes at presentation progress to more severe or persistent disorders. Consequently, this cohort should be the focus of active secondary prevention trials. In clinical practice, we are moving to combine the staging framework with likely pathophysiological paths (e.g. neurodevelopmental-psychotic, anxiety-depression, circadian-bipolar) to underpin enhanced treatment selection.

A fundamental aim of diagnosis is to guide treatment planning and predict illness course. Yet for too long psychiatric diagnosis, grounded on traditional silo-based approaches, has lacked clinical utility. This chapter explores the purpose of diagnosis and classification as well as the inability to validate diagnosis in psychiatry. It is proposed that new testable models are needed to improve the utility of diagnosis and support more personalised and sequential treatment selection. A number of new approaches have been put forward, including hierarchical and network-based methods, however at present, these offer limited value in guiding treatment selection. Clinical staging offers a viable solution. Clinical staging in psychiatry recognises that mental disorders are not static and discretely defined entities, but rather they are syndromes that overlap and develop in stages. The model ensures that interventions are proportional to both need and the risk of progressing to later stages and more established syndromes, which are likely to be comorbid, persistent, recurrent and disabling. Ultimately, it advocates a transdiagnostic approach to intervention, with a pre-emptive focus, that is based on risk-benefit considerations and patient needs. Clinical staging also provides a framework in which underlying biological mechanisms can be linked to each stage, to build a personalised and pre-emptive psychiatry.