Psychosis‐risk criteria in the general population: frequent misinterpretations and current evidence

Research on early detection and intervention in psychosis is going into its third decade. Its results are increasingly transferred into the clinic1, 2, and a tentative syndrome modeled on the description of “attenuated psychotic symptoms” (APS) in the ultra-high risk criteria has been included in the Section III (“Conditions for further study”) of the DSM-5. A wealth of evidence suggests: a) that symptomatic psychosis-risk criteria – in particular the APS and “brief intermittent psychotic symptoms” (BIPS) criteria, as well as the basic symptom criterion “cognitive disturbances” (COGDIS) – are associated with a significantly increased risk for psychosis in clinical samples, even in comparison with patients without psychosis-risk criteria from the same services1, and b) that specific psychological and pharmacological interventions reduce conversion rates to psychosis in adult patients with psychosis-risk criteria relative to control conditions and improve psychosocial functioning, although not to a significantly larger degree than control conditions2. Nevertheless, this indicated preventive approach continues to be criticized when considered through the lens of epidemiological findings3. Thereby, the focus is predominately on the APS criterion4. A main reason for the disparity between clinically and epidemiologically based viewpoints obviously originates from differences in assessments. Until recently, epidemiological studies have mainly used self-rating questionnaires or standardized lay-person interviews for the assessment of “psychotic-like experiences” (PLEs)4, 5, that are usually equaled to (attenuated) psychotic symptoms by critics3. Yet, questionnaire studies significantly overestimate the prevalence of PLEs already in comparison to lay-person interview studies5, and even more in comparison to clinician-based evaluations of APS using psychosis-risk assessment instruments6. Thus, PLEs are not an adequate proxy for APS/BIPS examined in clinical studies, and the conclusion – based on a wrongly assumed equality of these phenomena – that psychotic experiences are “a transdiagnostic dimension of psychopathology” and “a marker for the severity of non-psychotic states”3 must be regarded as unfounded when related to APS/BIPS assessed in clinical psychosis-risk research. Moreover, epidemiological studies usually assess the presence of PLEs but not their course or frequency, thus ignoring crucial requirements of APS/BIPS criteria. Indeed, the first sufficiently representative general population study (N=2,683) of young adults (age 16-40) in whom psychosis-risk criteria4 were assessed by trained clinicians – using established early detection instruments – documented that, although 11.96% (N=321) reported any lifetime and 7.53% (N=202) any current APS/BIPS, only 0.56% (N=15) fulfilled APS criteria (including onset or worsening within the past 12 months and at least weekly occurrence in the past month). One person meeting APS criteria also fulfilled BIPS criteria (including onset within the past three months and at least monthly occurrence for several minutes)4. Thus, the fact that (attenuated) psychotic phenomena may occasionally occur in persons of the general population with no past or present psychotic disorder does not rule out the significance of APS criteria as denoting a distinct and rather rare syndrome if additional course and frequency requirements are met. Such requirements are commonly part of the definition of mental disorders, when they are based on phenomena that might as well occur occasionally in everyday life, such as feeling low, sad and hopeless, or euphoric, or being afraid. Another feature that psychosis-risk criteria share with mental disorders is their frequent co-occurrence with other disorders, in particular depressive and anxiety disorders, with a prevalence of 23-94% in clinical1, 2 and 45% in general population samples4. Arguing that APS are solely a marker of the severity of non-psychotic disorders3 disregards the fact that depression and anxiety often arise from other mental or somatic disorders7, including non-affective psychoses8. In fact, a review of psychiatric comorbidity across different stages of schizophrenia confirmed the frequent co-occurrence of anxiety and depressive disorders throughout the course of the illness, including the prodrome8. From this, it was concluded that depressive and certain anxiety symptoms are “intrinsic to the illness and import a poorer outcome”8. Thus, the co-occurrence of psychosis-risk criteria and other mental disorders might indeed be “summarized as baseline differences in the severity of multidimensional psychopathology”3, yet with depression and/or anxiety rather than the far more infrequent APS/BIPS serving as transdiagnostic markers of severity. Longitudinal community studies using valid psychopathological assessments that are well comparable with clinical ones are needed to shed more light on the interplay and sequence of psychosis-risk criteria and non-psychotic mental disorders over time. Notably, in the critique of the psychosis-risk approach from an epidemiological viewpoint3, the basic symptom approach – in particular COGDIS – that next to APS and BIPS was recently recommended for the assessment of psychosis risk in the context of a guidance paper of the European Psychiatric Association1 – usually remains unmentioned. COGDIS and APS are equally common in clinical1, 9 and community samples4, often co-occur and, together, convey a much increased risk for psychosis, but not for other mental disorders, in clinical samples9. Yet, for their distinct quality1, 4, 10, cognitive basic symptoms cannot be referred to as “low-grade psychotic symptoms” or “psychotic experiences”3 and, consequently, critiques based on findings on PLEs cannot be extended to them. To conclude, while it is undisputed that more epidemiological as well as clinical research is needed to further improve prediction and prevention of psychosis1, 2 and to disentangle the dynamic relationship between psychosis-risk criteria and mental disorders, much of the recent critique of the psychosis-risk approach3 reflects prejudice and misperceptions of epidemiological but also of clinical findings and is not in agreement with current evidence. Frauke Schultze-Lutter1,2, Joachim Klosterkötter3, Wolfgang Gaebel2, Stefanie J. Schmidt1,3 1University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern, Bern, Switzerland; 2Department of Psychiatry and Psychotherapy, Medical Faculty, Heinrich-Heine University, Düsseldorf, Germany; 3Department of Psychiatry and Psychotherapy, University of Cologne, Cologne, Germany