Self-Reported Psychotic-Like Experiences Are a Poor Estimate of Clinician-Rated Attenuated and Frank Delusions and Hallucinations

Abstract

Background: One reason for the decision to delay the introduction of an Attenuated Psychosis Syndrome in the main text of the fifth edition of the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders was the concern that attenuated psychotic symptoms (APS) might in fact be common features in adolescents and young adults from the general population of no psychopathological significance in themselves. This concern was based on reports of high prevalence rates of psychotic-like experiences (PLEs) in the general population and the assumption that PLEs are a good estimate of APS. Although the criterion validity of self-reported PLEs had already been studied with respect to clinician-rated psychotic symptoms and found insufficient, it had been argued that PLEs might in fact be more comparable with mild, subclinical expressions of psychotic symptoms and, therefore, with APS. The present paper is the first to specifically study this assumption. Sampling and Methods: The sample consisted of 123 persons seeking help at a service for the early detection of psychosis, of whom 54 had an at-risk mental state or psychosis, 55 had a nonpsychotic mental disorder and 14 had no full-blown mental disorder. PLEs were assessed with the Peters Delusion Inventory and the revised Launay-Slade Hallucination Scale, and psychotic symptoms and APS were assessed with the Structured Interview for Prodromal Syndromes. Results: At a level of agreement between the presence of any PLE (in 98.4% of patients) and any APS (in 40.7%) just exceeding chance (κ = 0.022), the criterion validity of PLEs for APS was insufficient. Even if additional qualifiers (high agreement or distress, preoccupation and conviction) were considered, PLEs (in 52.8%) still tended to significantly overestimate APS, and agreement was only fair (κ = 0.340). Furthermore, the group effect on PLE prevalence was, at most, moderate (Cramer's V = 0.382). Conclusions: The prevalence of APS cannot be deduced from studies of PLEs. Thus, the high population prevalence rate of PLEs does not allow the conclusion that APS are common features of no pathological significance and would lack clinical validity as an Attenuated Psychosis Syndrome in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition. Rather, the population prevalence rate of APS has to be assumed to be largely unknown at present but is likely lower than indicated by epidemiological studies of PLEs. Therefore, dedicated studies are warranted, in which APS are assessed in a way that equates to their clinical evaluation.

Convergent and concurrent validity of the Frankfurt Complaint Questionnaire as a screener for psychosis risk

“…The sample size in this study was comparable to other screening evaluation studies in clinical samples (Kline and Schiffman 2014;Schultze-Lutter et al 2014) and was also sufficient to detect a clinically useful κ against the null hypothesis of no agreement between assessment modes with a power of 90% (Sim and Wright 2005). A clear disadvantage of the current study was, however, shown in the limited content validity of the FCQ as a result of the incomplete overlap between the BS included in FCQ and SPI-A/SPI-CY.…”

“…Although BS rely less on interviewer appraisal than attenuated positive symptoms, the FCQ's concurrent validity is only as good as that reported among self-report assessments of attenuated positive symptoms when a cut-off is chosen that optimises both sensitivity and specificity (Kline and Schiffman 2014). Thus, in line with reports on differences between self-report and clinical assessments of BS and attenuated positive symptoms (Granö et al 2011;Maß et al 1997;Ochoa et al 2008;Schultze-Lutter et al 2014;Yung et al 2009), we found a significant overestimation of clinician-rated BS by self-rated BS.…”

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

“…A recent study of an asthmamonitoring app reported that, while over 49,000 people downloaded the study's app, only 175 (0.35%) were actively engaged with it at six months 5 . A successful study of an app for alcohol use disorder 6 reported less success when later deploying without broad incentives 7 .…”

“…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 critics 3 . Yet, questionnaire studies significantly overestimate the prevalence of PLEs already in comparison to lay-person interview studies 5 , and even more in comparison to clinician-based evaluations of APS using psychosis-risk assessment instruments 6 .…”

Estados clínicos de alto riesgo para esquizofrenia y otras formas de psicosis: una breve revisión.

“…Investigaciones realizadas con éste y otros cuestionarios autoadministrados, demuestran que tienden a sobrevalorar los síntomas en las poblaciones estudiadas (89)(90)(91).…”