Adults with schizophrenia who threaten or attempt suicide have sharply increased risks of becoming violent, according to a recently published analysis.
Katrina Witt, a doctoral candidate affiliated with the University of Oxford (England), and her associates analyzed longitudinal data from the National Institute of Mental Health’s CATIE (Clinical Antipsychotic Trials of Intervention Effectiveness), a randomized controlled trial of antipsychotic medication in 1,460 adults with schizophrenia of generally moderate severity who were receiving usual care.
During a median follow-up of 15.7 months, 33.7% of the patients experienced suicidal ideation, 11.1% threatened suicide, and 5.8% attempted suicide, Ms. Witt and her associates reported (Schizophr. Res. 2014;154:61-7). About 8.3% of the patients showed violent behavior at some time as ascertained from interviews with family members.
In univariate analyses, suicidal threats and suicide attempts were significantly associated with violent behavior in both sexes, whereas suicidal ideation was not significantly associated for either sex.
In multivariate analyses that adjusted for a variety of comorbidities (alcohol misuse, drug misuse, diagnosed major depressive disorder, or diagnosed antisocial personality disorder), men and women had significantly elevated risks of violence if they made suicidal threats (hazard ratios, 3.8 and 9.4) or attempted suicide (hazard ratios, 2.8 and 4.4).
Additionally, for both sexes, the risks were elevated by roughly the same extent after adjustment for age or baseline scores for depression, hostility, positive symptoms, or poor impulse control. In women, adjustment for 6-month scores on these measures also made little difference; however, in men, adjustment abolished the significant association between suicide attempts and subsequent violence.
Of the three suicidality measures, suicidal threats yielded the greatest improvement in the prediction of violence for both sexes when added to a baseline risk model consisting of age, comorbid substance use disorder, and previous violence.
Ms. Witt and her associates cited several limitations. First, randomized controlled trials of antipsychotic effectiveness are "less likely to recruit individuals reporting thoughts of suicidality and self-harm." In light of that fact, it might not be possible to generalize the results of this study to all patients with schizophrenia.
Also, the CATIE data were not collected to meet the aims of this study, and as a result, it was not possible to include relevant confounding factors such as intelligence scores and "neighborhood socioeconomic deprivation."
Nevertheless, they said, their findings have implications for clinical care and for possible explanatory mechanisms.
"First, as part of the clinical risk assessment of violence in schizophrenia, as recommended by clinical guidelines in both the [United States] and [United Kingdom], a careful examination of history of suicidality should be included," they wrote.
"Second, the association between suicidal attempts and violence may be modified by 6-month depression, hostility, positive symptomatology, and poor impulse control scores in males. Given that medication adherence was monitored during the CATIE trial, this finding may suggest that acute symptomatology, perhaps exacerbated by medication nonadherence, may account for some of the association between suicidality and violence in males," they maintained. Thus efforts to ensure adherence might improve outcomes.
The investigators disclosed no conflicts of interest. The study was funded in part by the Wellcome Trust.