M-CHAT may often miss ASD

Researchers in Sweden and the United States report that the Modified Checklist for Autism in Toddlers (M-CHAT), commonly used by clinicians to screen children for autism spectrum disorders (ASD), misses a significant number of cases in children with medical conditions that increase the odds of ASD.

Benjamin Lassebro and colleagues analyzed data on 2,178 children at elevated likelihood of having ASD. The children in this group were extremely preterm or small for gestation age, had known brain damage, or had experienced neonatal encephalopathy or other severe neonatal medical problems. The researchers compared M-CHAT results with later ASD diagnoses, and found that the M-CHAT had 91% accuracy in ruling out an ASD diagnosis but only 62% accuracy in identifying true cases.

Study coauthor Ulrika Ådén comments, “The results show that M-CHAT works relatively well to rule out autism, but that it does not catch all children who later receive a diagnosis.” As a result, she says, clinicians should use additional tools to assess children with medical conditions that increase the likelihood of autism.

Citations

“Modified Checklist for Autism in Toddlers in a neonatal high-risk population,” Benjamin Lassebro, Matilda Morin, Weiyao Yin, Sven Sandin, and Ulrika Ådén, JAMA Network Open, March 27, 2026 (free online). Address: Benjamin Lassebro, Department of Biomedical and Clinical Sciences, Linköping University, Campus US, Linköping 58183, Sweden, [email protected].

—and—

“Study finds M-CHAT autism screening misses 38% of high-risk toddlers,” news release, Karolinska Institutet, March 27, 2026.