The average child with autism is 18 months old before his parents first begin to be concerned. Given the importance of early intervention, it’s crucial that parents and doctors both catch on to the symptoms as soon as possible.

A study published 28 May in European Child and Adolescent Psychiatry finds that two autism screens are better than one at identifying toddlers who need specialized clinical services.

Screens are used not to diagnose autism, but rather to identify children who need more specialized attention — for example, from a child psychiatrist or a behavioral therapist.

Two such screens, the Modified Checklist for Autism in Toddlers (M-CHAT) and the Early Screening of Autistic Traits (ESAT), are widely used to assess children between 18 months and 3 years of age during routine doctor visits.

Each screen asks parents yes-or-no questions about their child’s behavior, such as whether the child makes eye contact or engages in imaginative play. More than three ‘No’ answers flags the child as being at risk for autism, and merits a referral to a specialist.

The researchers looked at surveys from nearly 13,000 mothers in the Norwegian Mother and Child Cohort Study, which follows children from birth through early childhood. The surveys asked the mothers about their children’s symptoms, and also whether a doctor had referred their child to a specialist by 18 months of age.

The survey revealed that the ESAT highlighted 0.5 percent of the children, or 71, as being at risk of autism, and the M-CHAT flagged 6.4 percent, or 826 children.

Used together, the two tests identified 52 children as being at risk for autism. Parents of these children reported more clinical and developmental concerns than parents of children who screened positive on the M-CHAT alone.

Though the M-CHAT flagged more children than the ESAT, doctors — who did not see the results of the screens — had referred just 10 percent of the 826 children to specialists compared with 30 percent of the ESAT group. Those in the ESAT group also have more behavioral, health and developmental problems than those in the M-CHAT group.

It’s possible that the M-CHAT misidentifies some children in the general population as being at risk of autism, the researchers say. This may help children with autism get the services they need, but may also cause unnecessary anxiety when the child does not have autism. By contrast, the ESAT may be too conservative, potentially missing children with the disorder.

Neither screen picked up another 103 children in the study who received a referral for specialized services. This suggests that even used together, the tests may miss some children at risk of autism. (The researchers did not report how many children in the study have autism.) 

Still, the two tests together flag more children who need these services than either test alone, the researchers note. They recommend combining the screens for routine use by pediatricians.