Abstract
BACKGROUND & OBJECTIVES: There is a paucity of trained professionals for the diagnosis of autism spectrum disorder (ASD), and a large number of cases go undetected and are diagnosed only during adolescence. There is no screening instrument specifically developed for screening of Indian population for ASD. This study was undertaken to develop a screening instrument to screen ASD in north Indian Hindi speaking population by multipurpose health workers. METHODS: A 37-item instrument in Hindi with dichotomous yes/no responses [Chandigarh Autism Screening Instrument (CASI)] was developed to be applied on children aged 1.5-10 yr. The instrument was pilot tested and then reliability and validity of this instrument were tested. The sample included children with intellectual disability (n=75), ASD (n=83), other developmental disorders (n=87) and typically developing children (n=160). RESULTS: Reliability, construct and content validity testing of the instrument were performed, and a score of 10 as cut-off had sensitivity of 89.16 per cent, specificity of 89.13 per cent, positive predictive value of 67.89 per cent and negative predictive value of 96.96 per cent. A shorter four-item version (CASI Bref) has also been developed with good sensitivity (73.49%) and specificity (90.68%) at a cut-off score of 2. INTERPRETATION & CONCLUSIONS: CASI was found to be a valid instrument for screening general Hindi speaking population of north India with adequate sensitivity and specificity.