Skip to main content
<p style="margin-right: 0px; margin-bottom: 8px; margin-left: 0px;">Well child screeners are often questionnaires that are filled out by parents or caregivers during well child visits with pediatricians. While their use is common practice in many primary care settings, occasionally, questions arise as to their utility. Specifically, are the measures useful and can you <em>really</em> tell if a child has autism at age two? The goal of this post is to describe the clinical utility of the M-CHAT-R/F and explain the importance of its use.&nbsp;</p> <div></div>

Child Development - Well Child Autism Screening

A young child plays pretend doctor with a stuffed bear.

Well Child Autism Screening: How Does it Work and Does it Really Matter?

By Nicholas Bond, MA, LEND Trainee in Psychology

Well child screeners are often questionnaires that are filled out by parents or caregivers during well child visits with pediatricians. While their use is common practice in many primary care settings, occasionally, questions arise as to their utility. Specifically, are the measures useful and can you really tell if a child has autism at age two? The goal of this post is to describe the clinical utility of the M-CHAT-R/F and explain the importance of its use. 

For starters, the M-CHAT-R/F is a 20-question screener for autism spectrum disorder. It is normally filled out by parents at 18 month and 24 month well child checkups. The screener takes about five minutes to complete and it asks a variety of questions about the child’s current development. For example, “If you point at something across the room, does your child look at it?” or “When you smile at your child, does he or she smile back at you?” If three or more symptoms of autism are endorsed (in the most basic form of scoring), it is recommended that the child’s pediatrician refer the patient for a comprehensive autism evaluation.

While the cutoff score of three may seem arbitrary, it is anything but that. Research on the M-CHAT-R/F has found that using a cutoff score of three results in a 91.1% sensitivity and a 95.5% specificity (Robins et al., 2014). Let me explain what this means to everyone who is not as nerdy as me and does not particularly enjoy statistics and test construction as much as I do. A 91.1% sensitivity means that when we look at all the kids that have autism in the population, a cutoff score of three accurately identifies 91.1% of them. Conversely, a specificity of 95.5% means that of all the kids that do not have autism, 95.5% of them score less than three on this measure. What this means is that these simple twenty questions are able to correctly categorize more than 9 out of 10 kids with or without autism at the ages of 18 or 24 months.

So why does this all matter? First, it is amazing that a simple set of 20 questions, which only takes five minutes to complete, can identify the majority of children that will be diagnosed with autism. While it is not perfect, it does very well for a self-report measure. This leads me to my second point; it has been shown that physician concerns alone have a sensitivity of 24%, meaning that without screeners physicians are only able to correctly identify one quarter of the children that will be diagnosed with autism. Finally, research has continually shown that earlier interventions lead to better outcomes for individuals with autism. Diagnosis around a child’s second birthday, following a flagged M-CHAT-R/F would occur about two years earlier than the national average, likely leading to better long-term outcomes for the child.

Default Author Avatar IUSM Logo
Author

Cristy James

Cristina James is the Data Coordinator, Associate Training Director, and Family Discipline Coordinator in the Department of Pediatrics, Division of Child Development at Indiana University School of Medicine. She has over 10 years of professional experience and a life-long lived experience in neurodevelopmental disorders which, combined with her analytical skills, allow her to effectively span across functions to help provide and improve many LEND outcomes.

The views expressed in this content represent the perspective and opinions of the author and may or may not represent the position of Indiana University School of Medicine.