The Modified Checklist for Autism in Toddlers: A Follow up Study Investigating the Early Detection of Autism Spectrum Disorders in a Low Risk Sample

Background: The American Academy of Pediatrics recommends autism specific screening at 18 and 24 months of age. The Modified Checklist for Autism in Toddlers (M-CHAT) is an autism-specific, caregiver-report screening instrument designed for children 16-30 months of age. 

Objectives: The purpose of the current project is to provide updated findings regarding the use of the M-CHAT as an autism-specific screening instrument in a population-based sample and to provide longitudinal data from a sub-sample of children screened with the M-CHAT at a follow up around age four. 

Methods: The M-CHAT and the M–CHAT Follow-Up Interview (FUI) were used to screen 18,989 children aged 16-30 months at pediatric well child visits; screen positive cases were evaluated and classified as ASD or non-ASD. Rescreening was completed at 42-54 months using the M-CHAT or its revision, the M-CHAT-R, the Social Communication Questionnaire, and a question about ASD concerns to detect missed cases; 3,053 of 5,571 eligible cases completed the rescreen (55%). Screen positive cases were evaluated and classified as ASD or non-ASD. 

Results: Of the 18,989 screened cases at Time 1, 92 ASD cases were detected (86 based on M-CHAT + FUI, 6 bypassed FUI based on high M-CHAT scores), along with 79 false positive cases (75 of which had other significant delays) and 6 missed cases; Time 1 Positive Predictive Value (PPV) was .54 for ASD and .98 for all developmental delays. The utility of the M-CHAT total score and critical score cutoffs was assessed in the sample of 92 true positives; 91 children (98.9%) obtained an initial M-CHAT total score of three or higher; one was identified by critical score only. At Time 1, 235 children received an M-CHAT total score of 6 or higher; of those children, 166 (70.6%) continued to screen positive after the M-CHAT Interview and required an evaluation, Of those, 118 received an evaluation, 72 were diagnosed with ASD, 43 had other developmental delays, and only 3 were typically developing. Time 2 PPV for M-CHAT + FUI was .57. There were 10 children who screened negative on the M-CHAT at Time 1 but were evaluated and diagnosed with Autistic Disorder or PDD-NOS at Time 2, producing an upper bound estimate of M-CHAT sensitivity of .85. 

Conclusions:  Results suggest that the M-CHAT is effective at identifying ASD in a low risk sample. A M-CHAT total score cutoff of three or higher identifies nearly all screen positive cases and, for ease of scoring, using only the M-CHAT total score cutoff is acceptable. Results indicate that 98% of children who screen positive on the M-CHAT + FUI present with developmental delays requiring intervention, indicating that screen positive cases require immediate referral for evaluation and early intervention. An M-CHAT total score of 6 can serve as an appropriate clinical cutoff and providers can bypass the M-CHAT Interview and refer immediately to evaluation/ early intervention if a child obtains a M-CHAT score of 6 or higher. Rescreening is ongoing and will be complete in 2013.