In a recent review article of Guloksuz & van Os (Reference Guloksuz and van Os2017) in Psychological Medicine, authors suggest that the concept of schizophrenia should be abandoned and a single umbrella disorder – psychosis spectrum disorder (PSD), including schizophrenia spectrum disorders, schizoaffective disorder and bipolar disorder, should be created. They also argue for the existence of a general transdiagnostic psychosis phenotype at both clinical and subclinical levels – coinciding with PSD – encompassing affective and non-affective symptom dimensions, including positive, negative and disorganization symptoms, depression and mania. Their model suggests that common psychotic-like experiences (PLEs) (which are simply ‘psychotic experiences’ for the authors) are subthreshold expressions of PSD in the general population.

In their review, authors rightly discuss some controversial aspects of past high-risk research including topics, such as outcome bias, inability to distinguish risk for illness onset and risk for poor outcome using case–control paradigms. Van Os and colleagues have conducted very productive and admirable work in the field of epidemiology of PLEs and psychoses over the years (Linscott & van Os, Reference Linscott and van Os2010; van Os et al. Reference van Os, Linscott, Myin-Germeys, Delespaul and Krabbendam2009; van Os & Reininghaus, Reference van Os and Reininghaus2016). However, in my opinion, the concept of a single transdiagnostic psychosis phenotype is not supported by the evidence and their conceptualization of PSD will bring more harm than benefit.

In general medicine, objective measures, such as blood tests, can be used to define the continuum of a medical illness along with a severity gradient including subclinical and latent forms of the illness. However, in the absence of biological markers, the validity of such approach is problematic for psychotic symptoms, particularly for subclinical presentations. Equating PLEs with subtle psychotic symptoms is a misleading approach as PLEs in general population is a very heterogeneous concept. PLEs reflect psychotic experiences only in some individuals; in others they reflect other mental phenomena (i.e. dissociation). What is called a general transdiagnostic psychosis phenotype is, in fact, an amalgam of different concepts and their separate transdiagnostic networks. Even in the clinical level, there are fundamental issues for the proposal of the authors. For example, bipolar disorder has significant but only a partial genetic overlap with schizophrenia (Lichtenstein et al. Reference Lichtenstein, Yip, Björk, Pawitan, Cannon, Sullivan and Hultman2009). Some but not most patients with bipolar disorder should be considered in the same spectrum with schizophrenia (Bora, Reference Bora2015).

Authors emphasize the limitations of using enriched samples of poor outcome in understanding the characteristics of a broader phenotype of psychosis including diagnostic categories with favourable outcomes. However, using extremely diluted samples for true psychosis would only make matters worse. We need to use enriched samples for the spectrum of true psychosis not a spectrum of everything. In fact, one of the most remarkable findings of research of authors’ own group is the findings such as co-occurrence of negative (and disorganized) and positive symptoms or co-occurrence hallucinations and delusions in predicting persistence of PLEs (Dominguez et al. Reference Dominguez, Saka, Lieb, Wittchen and van Os2010; Smeets et al. Reference Smeets, Lataster, van Winkel, de Graaf, Ten Have and van Os2013). This is mainly an enrichment strategy for true positives than being an enrichment method for the poor outcome as authors suggest.

Replacing schizophrenia with an ill-defined PSD spectrum, which is, in fact, a mixture of spectrums of psychotic and non-psychotic conditions will not have much benefits in clinical practice and research and would have several harmful consequences. First, neurobiological and genetic research in these heterogeneous and diluted (for psychosis) samples would not give much insight about psychosis. Second, by labelling individuals presenting with PLEs (the milder end of the PSD spectrum) will lead to unnecessary stigma and stress associated with having subclinical psychosis among many who actually have non-psychotic conditions. We cannot effectively fight against pessimistic views and iatrogenic hopelessness associated with schizophrenia by further mystifying psychotic disorders by creating a unified construct including everything under the sun. Finally, authors advocate for use of general non-specific early interventions targeting transdiagnostic PSD spectrum, similar to the approach of the Headspace initiative (McGorry et al. Reference McGorry, Hamilton, Goldstone and Rickwood2016). In fact, interventions for the ill-defined transdiagnostic subclinical PSD in youth is not much different from interventions for a category of ‘youth with distress’. While there is no doubt that non-specific and general interventions for mild psychiatric presentations in youth (in any age group) might be helpful, accepting this approach as the ultimate goal would hamper the efforts to develop disorder-specific effective and true early intervention strategies for modifying the outcome of psychotic disorders (Bora, Reference Bora2017).

I agree with authors that the current concept of schizophrenia will die overtime. We will identify a number of disorders under the umbrella of schizophrenia-spectrum disorders and understand the subclinical continuum of these conditions. However, PSD spectrum concept based on an ill-defined general transdiagnostic PLEs construct would be a step backward in achieving this goal. We cannot recognize the individuals who are at the less severe end of the true psychosis spectrum and differentiate non-psychotic conditions presenting as PLEs with a purely phenomenological approach. Renaming and lumping together solely on phenomenological grounds would not get us closer to our common goals of modernizing psychiatry and improving outcomes of severe mental disorders.