(PS9-A3) Improving Suicide Assessment: Non-traditional Administration of the Columbia-suicide Severity Rating Scale

Despite being a widely used measure of suicidal thoughts and behaviors, the Columbia-Suicide Severity Rating Scale (C-SSRS) fails to capture the full spectrum of suicidal ideation (SI). Specifically, the C-SSRS navigation instructions assume that there is incremental severity in subtypes of SI (e.g., wish to be dead [passive SI] vs. thoughts of suicide [active SI]; Giddens et al., 2014), which is not uniformly supported by prior studies (e.g., Baca-Garcia et al., 2011) and may result in missed profiles of suicidal thinking. While past research has demonstrated discrepant reporting of suicide attempts across assessment types (e.g., single- vs. multi-item; Hom et al., 2016), to our knowledge, no study has examined differences in SI endorsement across C-SSRS administration formats. The present study retrospectively compared C-SSRS outcomes from asking all five SI subscale items to those that would have resulted from the standard administration of the measure (i.e., only asking about method, intent, or plan if non-specific active SI is endorsed). Towards this aim, 435 adult psychiatric inpatients (M_age = 40.3, 50.8% female) completed the C-SSRS during hospitalization. Consistent with previous conceptual work (Giddens et al., 2014), descriptive results derived from tabulating participants' response patterns revealed four distinct profiles of SI that are overlooked with the standard administration of the C-SSRS. Namely, 11 individuals would have been misclassified if the C-SSRS was conducted using the traditional format. These profiles included those with passive but not general active SI and thoughts of a method, plan, or intent. Further, the number of individuals reporting SI with a method significantly differed between administration methods, p = .002. When assessed comprehensively, a greater number of suicidal thoughts with a method are detected (n = 376 vs. 366). Thus, ceasing the C-SSRS administration if active SI is not endorsed results in a loss of information that may be important for determining the risk for suicide attempts (Linthicum & Ribeiro, 2022). Incorporating additional questions about SI may facilitate a better understanding among respondents (Hom et al., 2016) without a significant increase in clinical burden. However, additional research on SI assessment format is needed, ultimately to improve risk assessment in research and practice.