Autism Screen – The standard screen lacks the majority of infants with autism

Quick checkBefore seeing a pediatrician, parents can answer a series of questions about developing their toddler as an autism screening test.

FatCamera / iStock

The most widely used autism screening tool forgets a large proportion of toddlers who will be diagnosed with the disease later, according to a study published today in pediatrics¹.

The test, called a modified autism checklist in young children (M-CHAT), identifies less than 40% of children with autism. And 85% of children on the screen did not have autism.

The results echo those of similar studies conducted in Norway and Malaysia and highlight the need for new screens for autism. They also inform the ongoing debate about the opportunity to screen all infants for autism: Missing an autistic child prevents the child from being treated early, but reporting a child without autism can cause unnecessary anxiety to parents.

Researchers say it is still too early to settle the debate.

"We do not want anyone to interpret this as a recommendation not to screen," said chief investigator Whitney Guthrie, clinical psychologist and scientist at the Children's Hospital of Philadelphia. "Our conclusion is that, basically, there is still a lot of work to be done. We need to identify other methods of detecting these children who are missed. "

Early diagnosis:

Guthrie and her colleagues examined the health records of nearly 26,000 children aged 16 to 26 months treated in the hospital's network of 31 clinics. The visits lasted 4.5 years and began in January 2011.

The M-CHAT asks caregivers 23 questions about whether or not their child's development, such as: "Does your child ever claim, for example, to talk on the phone, take care of a doll or to pretend to something else? examined children's M-CHAT scores, as well as their medical records between 4 and 8 years of age.

Of the 20,375 children with medical records available, 454 had been diagnosed with autism at the age of 8 years. However, the M-CHAT missed – or gave a "false negative" result to – 61% of these children. And among the children that the screen detected as a whole, only 15% were subsequently diagnosed with autism, which means that the test also has a high "false positive" rate. The researchers presented the preliminary results of the study at the International Meeting on Autism Research 2017.

Most studies suggest that a positive M-CHAT result accurately predicts that autism does not follow up to the vast majority of children screened for negative.

The new study gives a better idea of ​​the sensitivity of the screen because the researchers also tracked these children, says Roald Øien, associate professor of special education and child development at UiT – the Arctic University of Norway in Tromsø – who did not participate in the study. .

However, autistic children with a positive result had been diagnosed 7.5 months earlier, on average, than those who had been missed by screening.

This gain in treatment time is an important reason for not ruling out the premise of universal screening, says Guthrie: "These children would lose those months of intervention sooner." End up getting the medical care they need.

More than 90% of the children had the screen. Children who have not been are more likely to be non-white and come from low-income families. The false positive rate in the M-CHAT group was significantly higher in these groups than in the high-income whites and higher in the girls than in the boys.

Watchful eye:

The effectiveness of M-CHAT depends on its use. The American Academy of Pediatrics recommends screening children twice – at 18 and 24 months – although it does not advocate any specific tool.

Only half of the children in the study were tested twice. In this group, the M-CAT did a little better: he reported half of the children diagnosed later with autism.

The high rate of false positives may also have reasons. Nearly 60% of infants whose score required a follow-up interview did not get one, perhaps because of time constraints; further tests may have removed these children from the positive group. "It speaks to the difficulty of this two-step screening in true primary care," Guthrie said.

If the M-CHAT is missing some groups or types of people with autism, new tests tailored to these groups could significantly improve the detection of the disease, explains Øien.

"The big question for the future is who is picked up by M-CHAT and who we miss," he says.

Yet, screening in toddlers offers the best chance of detecting autism early, according to other experts.

Pediatricians should be aware that M-CHAT is missing many autistic children, keep a watchful eye even on children who are screened negative and take careers' concerns seriously, says Lonnie Zwaigenbaum, a professor of pediatrics at the M-CHAT. University of Alberta in Edmonton. Canada. Zwaigenbaum did not participate in the study but co-wrote a comment on it.².

Tools that do not rely on parent observations could also help improve screen accuracy. Guthrie helps develop a technique involving computer vision that analyzes interactions between parents and young children. The method is designed to identify the characteristic traits of autism in these exchanges, in addition to traditional tests.