Examination of the properties of the Modified Checklist for Autism in Toddlers (M-CHAT) in a population sample.

Yama et al. (2012) checked if the M-CHAT still works when you screen toddlers as old as 48 months. They also asked whether a clinician can read the parent’s written answers instead of making a follow-up phone call.

The study used a survey design. Parents filled out the M-CHAT and wrote short notes about any “fail” items. Clinicians then scored the forms without calling anyone.

The M-CHAT stayed valid across the whole 20- to 48-month window. Skipping the phone call did not hurt accuracy as long as a clinician looked at the parent’s notes.

In short, you can save time by dropping the follow-up call if a trained eye reviews the written explanations.

Kamio et al. (2014) later tested the same tool at 18-month check-ups in Japan. They saw only modest sensitivity—about half of ASD cases were missed—so they kept the phone follow-up. The difference is age: 18-month-olds may need the call, while 20- to 48-month-olds do not.

Scarpa et al. (2013) found the opposite problem: in rural, low-income families the M-CHAT threw out far too many false positives. That looks like a clash, but it isn’t. Brie’s sample was general population; Angela’s was mostly minority caregivers with less schooling. Culture and income change how parents read the questions, not the tool itself.

Christopher et al. (2021) added another wrinkle. Toddlers with lots of tantrums or hyperactivity were harder to classify no matter what follow-up you used. So even with Brie’s shortcut, you still need clinical judgment when behavior is complex.

If you screen children between 20 and 48 months, you can drop the phone interview and let a clinician score the form with the parent’s notes. This cuts staff time and still keeps accuracy. Just remember: for younger toddlers, rural low-SES families, or kids with big behavior problems, the old two-step process—or extra clinical care—may be safer.