(PS2-44) A Systematic Review of Measures of Suicidality for Autistic People

Abstract Text: A relationship between higher rates of suicidality, characterized by suicidal ideation and behaviors, and diagnosis of autism spectrum disorder has been demonstrated within the last several decades (Blanchard et al., 2021; Demirkaya et al. 2016; Tsai et al. 2023). Yet, there is an absence of a validated questionnaire or measure used to assess suicidality in autistic people (Howe et al., 2020; Cassidy et al., 2020). However, progress has been made in recent years in the creation and modification of appropriate measures (Cassidy et al., 2021; Hedley et al. 2023). The current systematic review examines the availability of validated suicidality measures for autistic people and the psychometric properties and relative merits of these measures (PROSPERO registration number CRD42024559308). Following a systematic search of 6 databases, 922 articles were identified, of which 457 were automatically removed as duplicates. Eight additional articles were manually removed, and 2 were added through reference searches of relevant systematic reviews. Of the 459 articles screened by title and abstract, 78 were included for full-text screening, of which 13 met the criteria for review inclusion and data extraction. The selection criteria are (1) an autistic sample that is ideally compared to an allistic sample, (2) investigation of suicidality, (3) usage of a standardized measure to assess suicidality, (4) demonstration or report of validity of the scale in the autistic sample, and (5) that the study is empirical. Quality assessments were conducted and concluded acceptable for the included articles with a modified version of the LEGEND: Evidence Appraisal of a Single Study tool (Cincinnati Children's Hospital Medical Center, 2024). Results indicate the presence of 2 thoroughly validated tools that are psychometrically supported in an autistic population. The Suicide Behaviors Questionnaire-Autism Spectrum Conditions (SBQ-ASC) has strong convergent validity (rs = 0.877 between the SBQ-ASC and the SBQ-R), divergent validity comparing the SBQ-ASC and SBQ-R with the Anxiety Scale for Autism-Adults, content validity (>.80), and internal consistency (α=.792) and the Suicidal Ideation Attributes Scale-Modified which demonstrates adequate construct validity (TLI = 0.921), internal reliability (ω = 0.93), divergent validity between the SIDAS-M and COVID-19 Impact Scale-8 items, and predictive validity. Other measures used include the Patient Health Questionnaire-9 (PHQ-9) and Suicide Behaviors Questionnaire-Revised (SBQ-R), which show elements of good and poor validity. While the PHQ-9 had good internal consistency, it has poor construct validity with non-invariance of item 9 that inquiries about suicidality. A limitation of this review is the exclusion of non-published or “grey literature” in the search strategy, as there was no method to obtain such literature without biasing the articles included. We propose further validation and dissemination of these tools and a future focus on the theoretical relationship between suicidality and suicide attempts for autistic and general populations.