Sensitivity and specificity of a 3-item direct observation test for the early detection of infantile autism

Abstract

Introduction: Early diagnosis of autistic spectrum disorder is difficult. The use of an observational test that focuses on the "core" clinical features of autism spectrum disorder with good sensitivity and specificity and that is within the reach of the first contact physician, pediatricians or even non-medical staff would help to detect the disorder and initiate the process of diagnosis and early therapeutic measures. Objectives: To determine the sensitivity and specificity of a 3-item direct observation test (TOD3) for the detection of autism in preschool children. Methods: Two groups were formed by convenience: Group 1, formed by 25 subjects with autism by DSM-5 or by ADOS-2. And Group 2, 25 healthy subjects with no history related to ASD and absence of neurological symptoms. MCHAT-ES, 3-items test (TOD3) and ADOS-2 were applied. Results: The direct observation test of 3 items obtained a sensitivity of 92% and specificity of 100% with a positive predictive value (PPV) of 100% and a negative predictive value (NPV) of 88%. The positive likelihood ratio (+ LR) was tending to infinity and the negative (-LR) was 0.08. Conclusions: The 3-items direct observation test has good sensitivity and highly specific specificity for easy access and application detection of autism. It can be a useful clinical instrument for the pediatrician or first contact physician.