Terisa P. Gabrielsen, PhD, Assistant Professor of School Psychology at Brigham Young University, specializes in autism spectrum disorder. She talked to Dr. Tony Nimeh of PracticeUpdate Primary Care about the benefits of early diagnosis and intervention in very young children suspected of having autism and how to identify the early signs, even in infants as young as 2 months.

M-CHAT

Dr. Nimeh: Dr. Gabrielsen, could you please discuss the M-CHAT, the Modified Checklist for Autism in Toddlers?

Dr. Gabrielsen: The M-CHAT is freely available on the Internet, which is one of the nicest things about it. It has a very large evidence base supporting it. It was developed from the Checklist for Autism in Toddlers tool, which was developed by British researchers as a little bit more extensive tool that includes some observational measures. The developers of the M-CHAT took some of those ideas and made them into a 20-question, yes/no questionnaire that’s designed for parents to complete in the waiting room while they’re waiting to go in to see the pediatrician. The newly revised version is called the M-CHAT-R/F, and includes follow-up questions.

The checklist administration is now morphing a little bit, and it’s often being completed online. Parents can access it directly and take the results to their pediatricians. Some pediatricians send it out as preparation for a well-child visit. Autism Speaks has an automated version of the M-CHAT on their website. There are some commercial entities that have an automated version of the M-CHAT, and it’s always free. No one ever charges for it. There is also a website that’s just the M-CHAT: M-CHAT.org. The answers to the 20 yes/no questions are scored based on an algorithm. For providers, the most useful website for adopting the M-CHAT-R/F is the author’s website (Diana Robins).

We’re looking for fail answers of 3 or more questions as indication of an initially positive screen. So, if the answers to 2 of these 20 questions are considered fail answers, meaning that’s not what we expect in a child in that age group, that’s okay. That’s not enough to cause concern. However, once the number of fail answers get to 3 and above, we want to take a more extensive look at that child. The researchers have developed a follow-up interview to dig a little bit deeper into the questions that get a fail answer, and that is also freely available on the website.

Let’s say a physician gives an M-CHAT screening. In some of the studies that we have run, we had the receptionist give the parent the M-CHAT at check-in, based on the child’s age. The parent completed the questionnaire—we had English and Spanish versions—and it was given to the medical assistant, and then could be scored. In some cases, it goes into the electronic medical record (EMR), so that the results are available for the physician before he or she sees the child. Then, if warranted, the physician, either on the EMR or on paper, can go through some of the more in-depth follow-up questions to determine if there is a concern in those areas.

If there remains a concern on at least 3 of those questions, the recommendation is that the child be evaluated by someone with autism expertise. The physician would say that there’s enough of a concern here that someone should take a more extensive look at the child’s development and would refer the patient to the local early intervention system. There are certainly private providers who do evaluations, but the standard recommendation for all children would be to start with early intervention system referrals; then the physician can consider additional referrals for evaluation and diagnosis.

One of the standard referrals when autism is suspected is to check hearing to make sure that the child is failing to respond because of a social deficit and not a hearing deficit. When we think about risk factors, the American Academy of Pediatrics has an algorithm in which a certain number of risk factors indicate referral of a child for evaluation, and a failed screen is one of those risk factors, as is someone expressing a concern about development. Having siblings who have autism is also a risk factor. Research has recently verified something we’ve known for a long time—siblings are at a greater risk for autism if autism is in the family—but we’re now finding out that they’re at greater risk for other developmental delays as well. Even if they don’t have autism, they’re at greater risk—for example, for a speech delay or a motor delay.

There’s just been such an explosion in the early identification research. We’re finding all kinds of signs that we weren’t aware of before, including motor delays in children with autism. A motor delay might be one of the very earliest signs that there is a developmental problem that may be autism. Having a formal way for the physician to look at all of those signs—motor delays, speech delays, and social delays—gives the him or her a lot of evidence that it is time for a referral for a more in-depth evaluation. Even if the physician’s impression is that the child is doing well, all those little bits of evidence should add up to indicate that a more in-depth look is warranted.

AAP recommendations

Dr. Nimeh: Would you like to discuss the American Academy of Pediatrics recommendations?

Dr. Gabrielsen: The AAP recommendations were originally published in Pediatrics in 2007.¹ They were reaffirmed several years later.^2-3 The recommendations state that physicians should perform surveillance, as they always do, and look at the risk factors.

• If there is a sibling with autism, that’s one risk factor.
• If there’s a parental concern, that’s another risk factor.
• If any other caregiver has expressed a concern—a preschool teacher, a grandparent, a concerned neighbor—that’s a risk factor.
• If the pediatrician has a concern, that could be a risk factor as well.

If there are two or more of among those four risk factors, the recommendation is to make a referral for an evaluation and to simultaneously make a referral for early intervention services, which do not require diagnosis. Make a referral for audiology to check hearing. Schedule a follow-up visit. If only one of those risk factors is a concern, and the patient is at least 18 months old, we want to do a screen and evaluate the child’s social communication skills. Even if there are no risk factors, we still want to do the screen at the 18-month visit. If the results are negative, we just go ahead with regular preventative care and screen again at 24 months.

If the results of the screening are positive after the follow-up interview questions, we go ahead with the referral. If the results of that screening are negative, even if there is a risk factor, we want to provide parental education and do a follow-up within 1 month. We don’t wait until 6 months have passed, but do a follow-up again within 1 month to see what has changed, and then revisit the referral decision at that point.

All children should be screened twice by their second birthday. The AAP screening algorithm¹ is based on the number of risk factors. The screening results and the number of risk factors are what tells a physician what the next step should be and whether the child should referred for a more extensive evaluation or not.