BLOG

Finding Opportunities to Improve Screening for Autism

July 6, 2020

The AAP and Bright Futures recommends screening for autism spectrum disorder (ASD) to facilitate the
start of early intervention services.

Lewis First, MD, MS, Editor in Chief, Pediatrics

Content License: FreeView
Article type: Pediatrics Blog

The AAP and Bright Futuresrecommends screening for autism

spectrum disorder (ASD) to facilitate the start of early
intervention services. The revised Modified Checklist for Autism
in Toddlers (M-CHAT-R) is now commonly used, with
recommendations to screen at 18- and 24-months of age. How
good is the M-CHAT? Under research study conditions, quite
good (10.1542/peds.2013-1813) but in real world settings, less
so—at least according to a new study by Carbone et al
(10.1542/peds.2019-2314). The authors of this study, being released online this month in our journal, looked
at children during 18 or 24 month health maintenance visits from 20 clinics within a healthcare system for
evidence of a positive M-CHAT screen and then looked at these children again from age 5 on to see if they
had been given the coded diagnosis of an ASD. The authors evaluated more than 36,000 toddlers, of whom
73% were screened and 1.4% subsequently diagnosed with ASD. The good news is that children who
screened positive were more likely to be diagnosed with ASD and were picked up earlier (38.5 vs. 50.6
months (p<.001) than children not screened. The bad news is that Hispanic children in this study were less
likely to be screened. In addition, the authors found that the positive predictive value was not as good (18%)
when the M-CHAT was used in real-world settings and that the sensitivity of the M-CHAT was only 33%,
suggesting that most children who subsequently are diagnosed with autism are missed by the M-CHAT.

So what can be done about this, and should the M-CHAT-R still be used? The answer, according to a
commentary by Drs. Kate Wallace and Whitney Guthrie (10.1542/peds.2020-1467) from the Children’s
Hospital of Philadelphia, is yes but with a caveat. The authors believe that the M-CHAT, when relied on as
the sole screening tool for diagnosing a child with ASD, is not enough and other strategies and solutions
need to be introduced if we are going to identify as many children as possible with an ASD. Interestingly, the
commentary authors had done their own validation of the M-CHAT with a different population of children and
with similar results. That study was published last year in our journal (REF 5 of commentary). Check out
both this interesting study and commentary and consider at least continuing to monitor for developmental
signs of ASD even if the M-CHAT is negative during the health maintenance visits you have with toddlers
and preschoolers annually.

A Statewide Tiered System for Screening and Diagnosis of Autism Spectrum Disorder
Accuracy of Autism Screening in a Large Pediatric Network
Autism Screening With Online Decision Support by Primary Care Pediatricians Aided by M-CHAT/F
Facebook
Instagram

Copyright © 2020 American Academy of Pediatrics