Revised M-CHAT Improves Autism Screening for Toddlers

Post date: Jan 13, 2014 7:44:59 PM

Pediatricians and parents urged to use new screening questionnaire; easier to complete with fewer false positives

The M-CHAT autism screening tool for toddlers includes yes/no questions about social behaviors. (photos by Diana Robins)

Pediatricians and parents should use a new and improved version of the M-CHAT autism screen for toddlers with confidence, according to a new study in the journal Pediatrics.

The Modified Checklist for Autism in Toddlers Revised with Follow-up (M-CHAT-R/F) improves on the original M-CHAT screening tool. The American Academy of Pediatrics has long called on physicians to screen all toddlers for autism at 18- and 24-month checkups.

Like its predecessor, the M-CHAT-R/F consists of yes/no questions about early social behaviors such as pretend play and pointing. The checklist doesn’t diagnose autism spectrum disorder (ASD). Rather, a high score indicates that a child should be referred for full diagnostic testing.

“We want to encourage parents and pediatricians to make sure they’re using the improved version of this important screening tool,” says Alycia Halladay, Autism Speaks senior director of environmental and clinical sciences. “It’s appropriate for parents to ask for it specifically at their child’s 18- and 24-month checkups as well as any time they have concerns about a toddler’s development.”

An electronic version of the M-CHAT-R/F is available on the Autism Speaks websitehere. Autism Speaks is currently funding a study to ensure the accuracy of this online screening tool, which is designed to be easily used by parents. A printable version of the M-CHAT-R/F can be downloaded from the website of the tool’s author, Diane Robins of Atlanta’s Georgia State University, here.

Addressing challenges to well-child screening

The new revision came in response to concerns that the original M-CHAT produced too many “false positives,” says Dr. Robins. This refers to children whose M-CHAT score indicated high risk for autism, but whose diagnostic evaluation showed they didn’t have the disorder.

In recent years, Dr. Robins and her co-authors addressed the issue of false positives by adding a follow-up interview. Administered by a healthcare professional, the short structured interview clarified certain answers on the checklist. However, relatively few pediatricians used it, perhaps due to time constraints.

The revised M-CHAT simplifies some questions to clarify results and drops others that proved problematic, Dr. Robins says. This shortened the checklist from 23 to 20 questions. In addition, the new version integrates the previously separate follow-up interview.

Overall, the revised screening tool takes about 15 minutes to complete – including less than 5 minutes for parents to complete the 20 initial yes/no questions. Only when a child scores as “at risk” on the initial checklist do parents go on to complete the follow-up interview, which can be given over the phone.

In the newly published study, healthcare providers used the M-CHAT-R/F to screen more than 16,000 toddlers at routine 18- and 24-month checkups. Approximately 7 percent scored high enough to proceed to the follow-up interview. (The vast majority of children score at low risk for autism with the initial 20 questions.) In all, around 2 percent of the children were flagged for a full evaluation by an autism specialist. Of these, approximately 95 percent received a diagnosis of ASD or another developmental disorder warranting early intervention.

The researchers expressed hope that the increased ease and accuracy of the test would encourage its broader use.

The importance of universal screening

“Screening for autism is an essential part of well-child care,” Dr. Halladay emphasizes. “Combined with monitoring of broader developmental milestones, autism-specific screening is a powerful tool to reduce the age of diagnosis.” Early diagnosis, in turn, is essential for children to receive early intervention that can maximize gains in social, communication and learning skills.

“No screening tool is perfect,” Dr. Halladay adds. “Even if this improved screen doesn’t flag a child at high risk, parents and healthcare providers who have persistent concerns should nonetheless seek a full evaluation – without hesitation.”

Importantly, the children in the M-CHAT-R/F study were diagnosed about two years earlier than the national median age for autism diagnosis. While ASD can be reliably diagnosed by age 2, the average age of diagnosis has remained stubbornly above 4 years in the United States.

Autism Speaks Early Access to Care program seeks to reduce the average age of autism diagnosis and increase access to high-quality early intervention, with an emphasis on meeting the needs of underserved communities. Read more about the program’s screening events and materials here.