A Preliminary Analysis of the Ages and Stages Questionnaire (ASQ) Compared to the Modified Checklist for Autism In Toddlers – Revised (M-CHAT-R)

Background: The American Academy of Pediatrics recommends that children have autism-specific screening in addition to general developmental screening at the ages of 18- and 24-months.   However, empirical data are lacking to determine whether general developmental screening for all children, followed by autism-specific screening for only the positive cases, can detect ASDs.

Objectives: To investigate the concordance between general and autism-specific screening in children with a suspected developmental delay or autism spectrum disorder at 18- and 24-months of age.

Methods: 580 children were screened with the M-CHAT-R as well as the ASQ at 18- or 24-months of age.   Of those, 32 children failed the M-CHAT-R; 21 failed the M-CHAT-R initially, but passed the M-CHAT-R follow-up interview (Group 1), while another 11 failed the M-CHAT-R and the M-CHAT-R follow-up interview (Group 2), indicating that these children have autism-specific concerns confirmed on parent interview.   The ASQ total scores in each domain: Communication, Gross Motor, Fine Motor, Problem Solving, and Personal-Social were assessed for developmental concerns and compared between groups.

Results: Independent samples t-tests revealed significant differences between Group 1 and Group 2 on the Communication, Fine Motor, and Personal-Social domains of the ASQ, but not on Gross Motor and Problem Solving.  Within Group 2, 46% of children (n=5) had clinically significant scores on the two domains of the ASQ that would suggest a possible Autism Spectrum Disorder (i.e., Communication and Personal-Social domains).  Within Group 2 there were clinically significant concerns for 82% (n=9) of children on the Communication domain, and 55% (n=6) on Personal-Social; 91% of the group (n=10) had concerns on at least one of these domains.  In Group 1, only 4.8% of children (n=1) had clinically significant concerns on both the Communication and Personal-Social ASQ domains.  In Group 1, 19% of children (n=4) had clinically significant concerns on the Communication domain, and 24% of children (n=5) had concerns on the Personal-Social domain; 38% (n=8) had concerns on at least one of these domains.

Conclusions: Preliminary analyses suggest that the ASQ can identify most children who have autism risk.  Further data analysis will analyze how many children with Personal-Social and Communication concerns on the ASQ do not screen positive for possible autism, as well as analyze discriminating items. The application of the current results is limited based on the small sample size, but overall the data suggest that a general developmental screener may serve as an efficient first stage screener for autism.  However, further autism-specific screening would be necessary to specify concerns and decrease false positive referrals for comprehensive autism evaluations.