Are people at risk of psychosis also at risk of suicide and self-harm? A systematic review and meta-analysis.

Peter J Taylor, Paul Hutton, Lisa Wood

Research output: Contribution to journalArticlepeer-review


BACKGROUND: Suicide and self-harm are prevalent in individuals diagnosed with psychotic disorders. However, less is known about the level of self-injurious thinking and behaviour in those individuals deemed to be at ultra-high risk (UHR) of developing psychosis, despite growing clinical interest in this population. This review provides a synthesis of the extant literature concerning the prevalence of self-harm and suicidality in the UHR population, and the predictors and correlates associated with these events.

METHOD: A search of electronic databases was undertaken by two independent reviewers. A meta-analysis of prevalence was undertaken for self-harm, suicidal ideation and behaviour. A narrative review was also undertaken of analyses examining predictors and correlates of self-harm and suicidality.

RESULTS: Twenty-one eligible studies were identified. The meta-analyses suggested a high prevalence of recent suicidal ideation (66%), lifetime self-harm (49%) and lifetime suicide attempts (18%). Co-morbid psychiatric problems, mood variability and a family history of psychiatric problems were among the factors associated with self-harm and suicide risk.

CONCLUSIONS: Results suggest that self-harm and suicidality are highly prevalent in the UHR population, with rates similar to those observed in samples with diagnosed psychotic disorders. Appropriate monitoring and managing of suicide risk will be important for services working with the UHR population. Further research in this area is urgently needed considering the high rates identified.
Original languageEnglish
Pages (from-to)911-926
JournalPsychological Medicine
Issue number5
Early online date9 Oct 2014
Publication statusPublished - 2015


Dive into the research topics of 'Are people at risk of psychosis also at risk of suicide and self-harm? A systematic review and meta-analysis.'. Together they form a unique fingerprint.

Cite this