Psychosis-predictive value of self-reported schizotypy in a clinical high-risk sample

[Correction Notice: An Erratum for this article was reported in Vol 125(7) of Journal of Abnormal Psychology (see record 2016-47529-004). In the article, there was an error in the Author Note. The affiliation of Daniela Hubl was incorrectly listed as "University Hospital of Child and Adolescent Psychiatry and Psychotherapy, University of Bern." It should have been listed as "University Hospital of Psychiatry and Psychotherapy, University of Bern." All versions of this article have been corrected.] Schizotypy is considered an indicator of psychosis-proneness and therefore, a precursor to schizophrenia-spectrum psychosis. In the early detection of psychosis, the widely used ultra high-risk criteria refer to the positive features of schizotypy and schizotypal personality disorders (SPD). In clinical high risk (CHR) samples, self-reported or clinically assessed SPD, notably the lack of close friends, has been suggested to facilitate the prediction of psychosis. In community samples, self-reported schizotypy has mainly been assessed psychometrically using the 4 Wisconsin Schizotypy Scales (WSS; Chapman, Chapman, Kwapil, Eckbald, & Zinser, 1994), and the positive schizotypy dimension was consistently predictive of psychosis (Debbané et al., 2015). However, psychometrically assessed schizotypy has not yet been studied as a potential predictor of psychosis in CHR samples. To bridge this gap, we studied the psychosis-predictive value of 3 of the WSSs and their association with CHR state in a clinical sample. One hundred 28 patients (23 ± 7 years; 81% considered CHR) from 2 early detection services were followed for 12 to 101 months. Within 48 months, 36 (28.1%) converted to psychosis. Only physical anhedonia was associated with CHR state, and high scores for physical anhedonia were predictive of conversion in conjunction with the CHR state. Physical anhedonia rather than positive schizotypy scales might separate future converters from nonconverters in clinical samples already presenting a phenomenologically more extreme range on the psychosis continuum. Given their reported psychosis-predictive value in nonclinical samples, psychometric schizotypy measures in general might be useful for the initial screening of psychosis-proneness in the community, whereas physical anhedonia might be particularly useful in CHR samples. (PsycINFO Database Record