METHODS: A validation study. Forward/back translations followed by cultural adaptation. Face validity, operational equivalence were assessed. Panel of Psychiatrists assessed semantics, conceptual, content/construct validity using external criticism. Final Sinhalese version was applied to 100 Sinhalese speaking 10-19yr olds referred to Psychiatry units of two Teaching Hospitals following a suicide attempt or with suicidal ideation (cases) excluding psychotic disorder, acute emergencies or special needs and compared with same aged adolescents on treatment for respiratory problems as controls. Discriminant ability was calculated using area under the Receiver Operating Curve (AUC). Reliability was calculated using Cronbach\'s alpha.
RESULTS: Sinhalese version showed satisfactory face/contents/construct validity and operational equivalence. Median suicide risk score was 18 and mean 17.99 (SD = 7.56) for cases, median = 4 and mean = 4.86 (SD = 2.41) for controls. AUC was 0.969 (SE = 0.11), indicating a high ability to discriminate moderate/high from low suicidal risk. Cut off value, 8.5 was lower than original tool at a sensitivity of 91% and a specificity of 91%. Cronbach\'s alpha was 0.811 indicating high reliability and consistency.
CONCLUSIONS: Sinhalese version of ASAP-20 is a successful screening instrument to detect high suicide risk among adolescents in clinical and non clinical settings.