Screening tools for clinical high risk for psychosis

The purpose of this article was to review existing screening instruments that could be used to identify individuals who may be at increased risk for psychosis and to determine the suitability of these instruments.

Medline (Ovid) and PubMed were searched for peer-reviewed articles published in English, which reported performance evaluation of screening instruments for symptoms of high risk for psychosis. The articles' titles, abstracts and, when necessary, full texts were read to filter them against the selection criteria. Citations within relevant articles were hand searched for other potentially eligible studies.

This selection strategy resulted in identifying 56 articles (including three articles available only in an abstract format) that reported performance evaluation of 17 screening instruments.

The sensitivity of these scales ranged from 67% to 100% and the specificity ranged from 39% to 100%. The positive predictive value was less precise with scores ranging from 24% to 100%, and the negative predictive value ranging from 58% to 100%. There were several scales that might be useful for screening for individuals who are at increased risk for developing psychosis; however, the majority of measures are underexplored with poor validation.