Cannabis use was one correlate of suicide risk in psychosis-risk youth

Unraveling Suicidal Ideation and Behaviors in a Global Cohort of Individuals at Clinical High-Risk for Psychosis: Multivariable Correlates and Subthreshold Psychotic Symptoms.

Acta psychiatrica Scandinavica • • Moderately Relevant
🤖

AI Summary

This observational study examined suicidal ideation and behaviors in 1,443 people aged 12–30 who were at clinical high risk for psychosis. Using the Columbia Suicide Severity Rating Scale and multivariable Elastic Net models, researchers tested whether demographic characteristics, general mental-health symptoms, subthreshold psychotic symptoms, and social functioning were associated with suicidal risk. The study also compared psychosis-like symptom severity across groups with different levels of suicidal risk.

Lifetime suicidal ideation was reported by 68% of participants, while 26% reported previous suicidal behaviors. The most consistent correlates included current depressive symptoms or a present/past depression diagnosis, female sex at birth, cannabis use, visual or gustatory abnormalities, guilt, and lower role functioning. Suicidal ideation was also associated with some differences in disorganized communication and negative symptoms, while suicidal behaviors were associated with schizotypal personality disorder. Core unusual thoughts and experiences did not contribute to the predictive models.

Overall, subthreshold psychotic symptoms increased across the suicidal-risk spectrum, but the study concluded that suicidal ideation and behaviors primarily reflected clinical factors outside core psychosis-risk symptoms. Because this was an association-based analysis, it cannot establish that cannabis use or any other factor caused suicidal thoughts or behaviors, nor can it determine whether cannabis changes suicide risk. This is an abstract-based summary; the abstract does not provide the full clinical context or establish treatment implications.

💡 Key Findings

1
In this clinical high-risk cohort, 68% reported lifetime suicidal ideation and 26% reported previous suicidal behaviors.
High
90%
2
Across multivariable models, cannabis use was among the most consistent correlates of suicidal ideation and behaviors, alongside depression, guilt, lower role functioning, and female sex at birth; these are associations, not evidence of causation.
High
82%
3
Unusual Thoughts and Experiences, a core subthreshold psychosis-risk symptom, did not contribute to any suicidal-risk model, suggesting that suicidal risk was not explained primarily by core psychosis-like symptoms.
High
80%
4
Overall subthreshold psychotic symptom severity increased across groups with increasing suicidal risk, although the reported association was modest and does not establish clinical causation.
Good
78%

📄 Original Abstract

INTRODUCTION: Suicidal ideation and behaviors (SIB) are common among individuals at clinical high risk for psychosis (CHR-P), but their relationship with the overall clinical profile remains unclear. This study aimed to identify sociodemographic/clinical factors associated with SIB in CHR-P individuals and to examine whether subthreshold psychotic symptom severity increases with suicidal risk. METHODS: Data from the AMP SCZ initiative included 1443 CHR-P participants aged 12-30 years. SIB were assessed using the Columbia Suicide Severity Rating Scale (C-SSRS). Three Elastic Net regression models examined associations between suicidal ideation severity, intensity, and behaviors using 54 variables spanning sociodemographics, general psychopathology, subthreshold psychotic symptoms, and socio-occupational functioning. Moreover, ANCOVA compared overall psychotic symptom severity across five groups defined by increasing severity on the C-SSRS. RESULTS: Sixty-eight percent of participants reported lifetime suicidal ideation, and 26% reported previous suicidal behaviors. Model performance was modest but significant (cross-validated ideation severity: R2 = 0.161, RMSE = 0.916; ideation intensity: R2 = 0.193, RMSE = 0.898; suicidal behavior: accuracy = 0.685; AUC = 0.675; p < 0.001). Across all models, concurrent depressive symptoms, present/past diagnosis of depression, female sex at birth, cannabis use, visual/gustatory abnormalities, guilt, and lower role functioning were the most consistent correlates of SIB. Ideation was associated with lower disorganized communication expression and lower negative symptoms, whereas behaviors were associated with schizotypal personality disorder. Other core CHR-P symptoms, including Unusual Thoughts and Experiences, did not contribute to any model. ANCOVA (p < 0.001, partial η2 = 0.041) indicated increasing overall subthreshold psychotic symptoms across the SIB spectrum. CONCLUSION: SIB in CHR-P individuals reflects a multidimensional clinical profile primarily outside of core psychosis-risk symptoms. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT05905003.

Explore More Research

Stay informed about the latest cannabis science.

Your stash, decoded.