Reply

We thank Dr Large for his thoughtful and rigorous commentary on our systematic review and meta-analysis examining the predictive significance of the Columbia Suicide Severity Rating Scale (C-SSRS).^{Reference Large1} We appreciate the opportunity to clarify several conceptual issues raised, particularly regarding the interpretation of the predictive performance of the C-SSRS in suicide research.

First, with respect to suicide mortality, we explicitly acknowledged in our review that there were insufficient studies to support a meta-analysis of suicide death outcomes. This limitation reflects the well-recognised methodological constraints inherent in studying events with very low base rates, even in high-risk clinical populations, and is not a limitation specific to the C-SSRS. The absence of robust meta-analytic evidence for suicide death should therefore not be interpreted as evidence against the instrument’s clinical utility, but rather as lack of evidence due to a statistical limitation of the low base rate of suicide death.

Second, we agree that the effect sizes observed for the prediction of non-fatal suicide attempts are modest. However, within the field of suicide research, odds ratios in the range of two to three are clinically meaningful^{Reference Franklin, Ribeiro, Fox, Bentley, Kleiman and Huang2} and are comparable to those associated with many accepted risk factors in medicine, such as smoking for acute myocardial infarction.^{Reference Teo, Ounpuu, Hawken, Pandey, Valentin and Hunt3} Importantly, our findings do not imply deterministic prediction at the individual level. Instead, they support the role of the C-SSRS as a probabilistic risk-stratification tool, capable of identifying groups of patients with an increased likelihood of future suicidal attempt relative to baseline risk.

Third, we concur that suicide risk assessment tools should not be used as mechanistic or ritualised substitutes for clinical judgement. As emphasised in our discussion, C-SSRS findings must be integrated within a comprehensive clinical assessment that includes dynamic clinical variables, protective factors and contextual considerations. The value of structured instruments such as the C-SSRS lies not in eliminating uncertainty, but in making the assessment of suicidal ideation and behaviour more systematic, transparent and comparable across clinicians and settings.^{Reference Malone, Szanto, Corbitt and Mann4}

Finally, although suicide death is the most tragic outcome, non-fatal suicide attempts represent a clinically significant end-point in their own right and remain the strongest known predictor of future suicide mortality.^{Reference Mann and Rizk5} From both clinical and public health perspectives, improving the identification of individuals at increased risk of suicide attempts is a necessary and actionable step in suicide prevention.

In summary, our findings show that using the C-SSRS provides modest but meaningful prognostic information that can support clinical decision-making when used appropriately and in conjunction with comprehensive care.