Suicidality in Psychosis-Risk Youth: Depression, Not Psychosis, Drives Risk

Large Global Study Finds Depression, Not Psychosis, Drives Suicidal Risk in At-Risk Youth

Suicidal ideation and behaviors (SIB) among youth at clinical high risk for psychosis (CHR-P) are more strongly associated with depression, cannabis use, and socio-occupational impairment than with core psychosis-risk symptoms, according to a major multinational study published September 2026 and registered under ClinicalTrials.gov identifier NCT05905003. The research, conducted as part of the Accelerating Medicines Partnership Schizophrenia (AMP SCZ) initiative, analyzed data from 1,443 participants aged 12–30 years across diverse global sites, using the Columbia Suicide Severity Rating Scale (C-SSRS) and advanced multivariable modeling.

Key findings indicate that 68% of CHR-P participants reported lifetime suicidal ideation and 26% reported prior suicidal behaviors. However, the severity and occurrence of these outcomes were predicted most consistently by depressive symptoms, a history of depression, female sex at birth, cannabis use, feelings of guilt, and lower role functioning. In contrast, classic psychosis-risk symptoms—such as unusual thoughts and experiences—did not significantly contribute to the models predicting suicidality.

Mechanisms: Multidimensional Risk Profile Beyond Psychosis Symptoms

Suicidality in youth at risk for psychosis is best explained by a multidimensional profile dominated by mood, substance use, and functional factors, rather than core psychotic symptoms. Elastic Net regression models, which included 54 sociodemographic and clinical variables, consistently identified depression-related variables and socio-occupational impairment as the strongest correlates of suicidal ideation and behaviors. Cannabis use and certain perceptual abnormalities (visual/gustatory) also emerged as significant, while negative symptoms and disorganized communication were inversely related to ideation.

Notably, the study’s ANCOVA analysis found only a modest increase in overall subthreshold psychotic symptoms across the spectrum of suicidality (partial η2 = 0.041), suggesting that while psychosis-risk symptoms may track with overall clinical severity, they are not primary drivers of suicide risk. This challenges longstanding assumptions that suicidality in this population is predominantly a function of emerging psychosis, and highlights the need for comprehensive assessment and intervention strategies.

Policy and Research Implications: Rethinking Intervention and Trial Design

Interventions targeting depression and functional impairment may be more effective in reducing suicide risk among CHR-P youth than those focused solely on preventing psychosis onset. For policymakers and clinical researchers, these findings underscore the importance of integrating mood disorder assessment and treatment—alongside substance use interventions—into early psychosis risk services.

• Clinical Trials: Psychedelic research and other experimental interventions in CHR-P populations should carefully screen for and address comorbid depression and cannabis use, as these factors may confound outcomes and safety profiles.
• Service Design: Early intervention programs should prioritize resources for mood and functional support, rather than restricting focus to psychosis prevention.
• Regulatory Considerations: Agencies reviewing trial protocols for novel therapies, including psychedelics, may require more robust risk mitigation for suicidality linked to depressive symptoms, not just psychosis progression.

An underappreciated implication is that excluding participants with significant depression or substance use from trials—common in psychedelic studies—may inadvertently exclude those at highest suicide risk, limiting real-world applicability and safety data.

Risks, Limitations, and Unknowns

The study’s models achieved only modest predictive power (e.g., R2 values around 0.16–0.19 for ideation severity), highlighting the complexity and multifactorial nature of suicidality in this population. While depression and functional impairment are key correlates, a substantial proportion of risk remains unexplained, and causal relationships cannot be definitively established from cross-sectional data.

Other limitations include potential site and cultural differences in reporting, the broad age range (12–30), and reliance on self-report measures for sensitive topics like suicidality and cannabis use. The findings may not generalize to all CHR-P populations, particularly those outside research settings or with different healthcare access.

For psychedelic researchers, a real failure mode is the risk of underestimating suicidality if trial designs focus too narrowly on psychosis symptoms, missing critical mood and substance use drivers. This can lead to both safety concerns and missed opportunities for effective intervention.

Looking Ahead: Integrating Findings Into Practice and Research

Future research should prioritize longitudinal designs to clarify the temporal relationships between depression, cannabis use, functional impairment, and suicidality in CHR-P youth. For clinical and trial settings, comprehensive risk assessment tools that capture mood, substance use, and social functioning are essential. Intervention development—including psychedelic-assisted therapies—should be tailored to address the full spectrum of risk factors, not just psychosis prevention.

Ultimately, this study marks a shift in how suicidality is understood and managed in youth at high risk for psychosis, with broad implications for clinical care, research priorities, and policy frameworks worldwide.

By Dr. Alex R. Bennett, PhD (Neuroscience, University of Toronto). Reviewed by Dr. Maya Lin, MD, FRCPC (Psychiatry) on 2026-09-12. We base our analysis on the original AMP SCZ study (OpenAlex W7212248045), ClinicalTrials.gov NCT05905003, and primary data as cited.