The Suicide Assessment Kit-Modified Interview: Development and preliminary validation of a modified clinical interview for the assessment of suicidal thoughts and behavior in autistic adults

The increased prevalence of suicide among autistic people highlights the need for validated clinical suicide screening and assessment instruments that are accessible and meet the unique language and communication needs of this population. We describe the preliminary preregistered psychometric validation of the Suicide Assessment Kit- Modified Interview, in a sample of 98 autistic adults (58% women, 34% men, 7% nonbinary; M (AGE) = 41.65, SD = 12.96). A four-item negative affect score derived from the Suicide Assessment Kit-Modified Interview demonstrated adequate reliability (omega = 0.796, BCa 95% confidence interval = [0.706, 0.857]), as well as good convergent validity with related measures. Ordinal Suicide Assessment Kit-Modified Interview "categories" (1-5) demonstrated divergent validity (rs = -0.067 to 0.081) and good convergent validity, strongly correlating with mental health (rs= 0.446 to 0.744) and suicide assessment instruments (= 0.576 to 0.696). Suicide Assessment Kit-Modified Interview categories showed strong ability to predict participants identified by clinicians as "above low risk" of future suicide attempt (area under the curve = 0.887, posterior Mdn = 0.889, 95% credible interval = [0.810, 0.954], P AUC > 0.8 = 0.976). Suicide Assessment Kit-Modified Interview classifications > " Category 3" provided an observed sensitivity of 0.750 ( Mdn = 0.810, [0.669, 0.948],P-se > 0.8 = 0.544) and an observed specificity of 0.895 ( Mdn = 0.899, [0.833, 0.956], P-Sp >( 0.8) = 0.995) for "above low risk" status. Our findings indicate that the Suicide Assessment Kit-Modified Interview is a psychometrically strong clinical assessment tool for suicidal behavior that can be validly administered to autistic adults without intellectual disability.