• Online Features  This Article  • Full text  • PDF  • eSupplement  • Reply to article  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Citation map  • Contact me when this article is cited  Related Content  • Similar articles in this journal  Topic Collections  • Neurology  • Cognitive Disorders  • Psychiatry  • Psychiatry, Other  • Alert me on articles by topic  Social Bookmarking   What's this?

Nomi Werbeloff, PhD; Marjan Drukker, PhD; Bruce P. Dohrenwend, PhD; Itzhak Levav, MD; Rinat Yoffe, MA; Jim van Os, MD, PhD; Michael Davidson, MD; Mark Weiser, MD

Arch Gen Psychiatry. Published online January 2, 2012. doi:10.1001/archgenpsychiatry.2011.1580

Context  It has been suggested that attenuated psychotic symptoms (APSs) reported by people who do not have psychotic disorders signal risk for later severe mental illness.

Objective  To investigate this suggestion using follow-up assessments of hospitalization for clinical diagnoses of nonaffective psychotic and other psychiatric disorders.

Design  Longitudinal cohort study of self-reported APSs with outcome assessment of severe mental illness obtained through linkage with a national hospitalization case registry.

Setting  Israel.

Participants  A stratified full probability sample of 4914 persons aged 25 to 34 years who were screened for psychopathology in the 1980s.

Main Outcome Measure  Subsequent psychiatric hospitalization was ascertained using the psychiatric hospitalization registry, with a mean follow-up of 24 years.

Results  After removing subjects with diagnosable psychotic disorders at baseline, 57.2% of the remaining sample reported at least 1 weak (infrequent) APS and 14.3% reported at least 1 strong (frequent) APS in the year preceding the assessment. Self-reported APSs predicted risk of later hospitalization for nonaffective psychotic disorders, mostly during the 5 years after baseline (adjusted odds ratio = 4.31; 95% CI, 2.21-8.41; positive predictive value = 1.27%; population attributable risk fraction = 33%). Also, APSs increased the risk of later hospitalization for other psychiatric disorders, albeit to a lesser extent (adjusted odds ratio = 2.21; 95% CI, 1.02-4.82).

Conclusions  Self-reported APSs signal risk for later nonaffective psychotic disorders but are not clinically useful as predictors. The difference between these population-based data and the high-risk literature in terms of the positive predictive value (1% vs 10%, respectively) and the time window of transition (5 years vs 12 months, respectively) can be attributed to the selective enrichment strategies that produce high-risk samples.

Author Affiliations: Department of Psychiatry, Sheba Medical Center, Tel Hashomer (Drs Werbeloff, Davidson, and Weiser), Division of Mental Health Services, Ministry of Health, Jerusalem (Dr Levav and Ms Yoffe), and Department of Psychiatry, Sackler School of Medicine, Tel Aviv University, Ramat Aviv (Drs Davidson and Weiser), Israel; Department of Psychiatry and Psychology, South Limburg Mental Health Research and Teaching Network, European Robotics Research Network, Maastricht University Medical Centre, Maastricht, the Netherlands (Drs Drukker and van Os); Department of Psychiatry, Mailman School of Public Health, Columbia University, New York State Psychiatric Institute, New York (Dr Dohrenwend); and Department of Psychosis Studies, Institute of Psychiatry, King's College London, King's Health Partners, London, England (Dr van Os).

CiteULike    Connotea    Delicious    Digg    Facebook    Reddit    Technorati    Twitter     What's this?