What makes the psychosis 'clinical high risk' state risky: psychosis itself or the co-presence of a non-psychotic disorder?

Laila Hasmi,Sinan Guloksuz,Michael C O'Donovan,Saskia Van Dorsselaer,Jim Van Os,Alexander Richards,Bochao D Lin,Maarten Bak,Gunter Kenis,Ron De Graaf,Lotta-Katrin Pries,Margreet Ten Have,Jurjen J Luykx,Bart P.F Rutten

doi:10.1017/s204579602100041x

Abstract

Although attenuated psychotic symptoms in the psychosis clinical high-risk state (CHR-P) almost always occur in the context of a non-psychotic disorder (NPD), NPD is considered an undesired 'comorbidity' epiphenomenon rather than an integral part of CHR-P itself. Prospective work, however, indicates that much more of the clinical psychosis incidence is attributable to prior mood and drug use disorders than to psychosis clinical high-risk states per se. In order to examine this conundrum, we analysed to what degree the 'risk' in CHR-P is indexed by co-present NPD rather than attenuated psychosis per se. We examined the incidence of early psychotic experiences (PE) with and without NPD (mood disorders, anxiety disorders, alcohol/drug use disorders), in a prospective general population cohort (n = 6123 at risk of incident PE at baseline). Four interview waves were conducted between 2007 and 2018 (NEMESIS-2). The incidence of PE, alone (PE-only) or with NPD (PE + NPD) was calculated, as were differential associations with schizophrenia polygenic risk score (PRS-Sz), environmental, demographical, clinical and cognitive factors. The incidence of PE + NPD (0.37%) was lower than the incidence of PE-only (1.04%), representing around a third of the total yearly incidence of PE. Incident PE + NPD was, in comparison with PE-only, differentially characterised by poor functioning, environmental risks, PRS-Sz, positive family history, prescription of antipsychotic medication and (mental) health service use. The risk in 'clinical high risk' states is mediated not by attenuated psychosis per se but specifically the combination of attenuated psychosis and NPD. CHR-P/APS research should be reconceptualised from a focus on attenuated psychotic symptoms with exclusion of non-psychotic DSM-disorders, as the 'pure' representation of a supposedly homotypic psychosis risk state, towards a focus on poor-outcome NPDs, characterised by a degree of psychosis admixture, on the pathway to psychotic disorder outcomes.

Highlights

There is considerable interest in the psychosis clinical High-Risk state (CHR-P) as a paradigm to identify the risk factors, mechanisms and early intervention potential associated with the onset of psychotic disorder (Fusar-Poli et al, 2020)
Recent prospective work showed that much more of the clinical psychosis incidence is attributable to prior mood and drug use disorders than to psychosis clinical high-risk states (Guloksuz et al, 2020)
We found that of the total incidence of psychotic experiences (PE), approximately 25% involved a combined phenotype of PE + non-psychotic disorder (NPD)

Summary

Introduction

There is considerable interest in the psychosis clinical High-Risk state (CHR-P) as a paradigm to identify the risk factors, mechanisms and early intervention potential associated with the onset of psychotic disorder (Fusar-Poli et al, 2020). CHR-P is defined mainly by the presence of attenuated psychotic symptoms which, represent the central part of the definition of attenuated psychosis syndrome (APS) in (the appendix of) DSM-5 (Tsuang et al, 2013). Some have proposed that the CHR-P/APS construct may be valid (Woods et al, 2009; Salazar de Pablo et al, 2020), others point to epistemic, conceptual and methodological limitations surrounding different aspects of CHR-P research (van Os and Guloksuz, 2017; Ajnakina et al, 2018; Moritz et al, 2019) Clarification of these issues is urgently required given increasingly large-scale research projects based on the CHR-P paradigm

Methods

Results

Conclusion