The burden of awareness of psychometric risk for schizophrenia Richard J. Linscott ⁎ , Fraser V. Cross Department of Psychology, University of Otago, P. O. Box 56, Dunedin 9054, New Zealand Received 18 February 2007; received in revised form 15 November 2007; accepted 3 March 2008 Abstract Participants in studies of psychometric risk for schizophrenia are rarely informed of their risk status. Nondisclosure may be justifiable if the harmful effects of disclosure outweigh its benefits. We examined whether disclosure may adversely affect well- being and, if so, factors that predict the degree of adverse effect. Undergraduates (n = 114) rated the anticipated impact—on felt distress, coping, optimism, helplessness, future lifestyle choices, and survival—of discovering they were at risk for schizophrenia and six other diseases. They also completed measures of potential predictors of this impact, including knowledge about schizophrenia, vicarious experience of schizophrenia, their potential to suffer stigmatization because of schizophrenia, and schizotypy. Participants judged schizophrenia risk more negatively than risk for heart disease, arthritis, depression, and diabetes, and less negatively than risk for cancer and Alzheimer's disease. Higher disorder-nonspecific impact, greater stigma, and lower psychometric risk for schizophrenia together provided the best linear prediction of schizophrenia-specific impact. Awareness of schizophrenia risk creates a significant adverse impact, the level of which may be greatest among those with lowest risk. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Disclosure; Ethics; Schizophrenia risk; Schizotypy; Screening; Stigma 1. Introduction General population screening for features of psycho- metric risk for schizophrenia (also called schizotypy , psychosis proneness, or subclinical psychosis) is based on the premise that the presence of such features indicates the presence of a vulnerability for schizo- phrenia or nonaffective psychosis (Meehl, 1990; van Os et al., 2000). Longitudinal research shows that the rate of nonaffective psychosis is higher among those who exhibit features of schizotypy than among those who do not (Chapman et al., 1994; Poulton et al., 2000). Despite or perhaps because of this link, screening often proceeds without research participants being informed of the nature of the construct researchers presume they are measuring. Rather, participants are invited to complete surveys of feelings or attitudes or some equally imprecise construct (Chapman et al., 1982; Venables et al., 1990; Korfine and Lenzenweger, 1995; van Os et al., 1999). Others have completed schizotypy measures embedded within scales that measure per- sonality (Golden and Meehl, 1979; Linscott et al., 2006) or anxiety (Nielsen and Petersen, 1976), or that make no reference to any particular construct (Claridge and Broks, 1984; Chapman and Chapman, 1989; Raine, 1991; Rawlings and MacFarlane, 1994). Although widespread, this practice raises several ethical questions Available online at www.sciencedirect.com Psychiatry Research 166 (2009) 184 – 191 www.elsevier.com/locate/psychres ⁎ Corresponding author. Tel.: +64 3 4795689; fax: +64 3 4798335. E-mail address: linscott@psy.otago.ac.nz (R.J. Linscott). 0165-1781/$ - see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2008.03.007