Quick summary: A large review of data from more than 6 million people across 13 countries has found that several psychiatric medications are associated with meaningfully lower risks of suicide and self-harm in people with serious mental health conditions. Clozapine, lithium, and certain antidepressants showed the strongest protective effects, while benzodiazepines were linked to higher suicide mortality across multiple diagnoses. The findings support a more personalised approach to prescribing, with medication choices treated as a core component of suicide prevention alongside psychological and social support.
A major new review has found that certain psychiatric medications are linked to a significantly lower risk of suicide and self-harm in people living with serious mental health conditions. The findings, drawn from data covering more than 6 million individuals across 13 countries, offer some of the clearest real-world evidence yet that the right medication, matched to the right diagnosis, may play a meaningful role in suicide prevention. The findings were published in eClinicalMedicine.
The research, led by psychiatrists at the University of Oxford, pooled results from 48 observational studies conducted over three decades. Professor Seena Fazel, professor of forensic psychiatry at the University of Oxford and one of the researchers behind the review, explained why this approach was taken: “For suicide-related outcomes, RCTs are often not feasible because they require long follow-up periods, very large sample sizes, and may raise ethical concerns. In addition, many trials exclude individuals at elevated suicide risk at baseline, limiting their generalisability to real-world clinical populations.”
For people with schizophrenia spectrum disorders, second-generation antipsychotic medications showed the most consistent protective effects. Clozapine was associated with a 60% reduction in suicide mortality, while olanzapine and quetiapine also showed meaningful reductions. Risperidone was linked to lower risks of both suicide attempts and suicide death.
In bipolar disorder, lithium emerged as one of the most protective treatments, with strong associations with reduced suicide risk across multiple studies. Valproic acid also showed a link with lower suicide mortality, though the researchers noted that current clinical guidelines caution against its use because of concerns over long-term side effects.
For people with depression, both selective serotonin reuptake inhibitors and older tricyclic antidepressants were associated with reduced risk of suicide death. The findings provide real-world support for the use of antidepressants in managing suicide risk in depressive illness, complementing and in some respects extending the evidence from clinical trials.
Not all findings were reassuring. Benzodiazepines, a class of medications commonly prescribed for anxiety and sleep disorders, were associated with a higher risk of suicide mortality across most diagnostic categories, including schizophrenia, bipolar disorder, and personality disorders. The authors suggest that people prescribed these medications should receive regular clinical reviews.
Professor Fazel pointed to a broader ambition for this line of research: “A major clinical challenge is linking prognostic models for suicide risk with effective interventions to enable a stratified medicine approach to suicide prevention. These models have demonstrated good external validity, but their clinical utility depends on linkage to effective interventions, as current approaches often apply broadly similar interventions across heterogeneous risk groups, reducing efficiency, potentially wasting resources, and diluting treatment effects.”
The researchers were careful to stress that the findings do not prove cause and effect, and publication bias was identified for some findings. Despite these limitations, the authors argue that personalising medication choices based on a patient’s specific diagnosis should be considered a central component of suicide prevention strategies, alongside psychological support and social interventions.
