How early can autism be detected in babies, and how soon can they be treated?
A baby's first birthday visit to the pediatrician usually includes a screening for the social deficits common with autism spectrum disorder. But doctors and scientists tend to agree that they can't make a very reliable diagnosis until the toddler is 2 years old. The bulk of treatment programs begin then.
Several recent studies, however, have documented subtle signs of the disorder, including erratic eye motion, among infants as young at 2 months old. That and other behavioral differences become more noticeable between 6 months and a year of age, other studies have shown.
Researchers at UC Davis' MIND Institute have been watching and playing with babies for many years to study autism and other development issues. This time, researchers set out to test whether they could accurately identify early signs associated with autism, and whether parents would be willing and able to follow directed therapy aimed at improving interaction with their babies. And if they did, would it help?
The study, published online Tuesday in the Journal of Autism and Development Disorders, suggests a qualified yes on all counts. The data from the pilot program jibe with a growing body of research that shows that guided social interaction between a mother and her baby can decrease autism symptoms by age 3.
The babies whose parents underwent the coaching showed far fewer autism-related measures at 3 than peers whose parents declined to participate, even though their infants met all criteria. Infants in the training program also scored higher than non-participants who later were diagnosed with autism, the study found.
"It gives us a little hint that the children may well have gone on to have more difficulties had we not done this intervention early," said lead investigator Sally J. Rogers, a developmental psychologist at the MIND Institute. "But it's only a hint, not proof."
The sample of the pilot study was small – just seven babies from UC Davis programs or the community underwent training, and babies were not assigned randomly to different treatments. So, results will have to withstand more rigorous testing.
In some ways, the study was as much a test for parents as for babies. Researchers wanted to know whether mothers could be professionally coached to maximize opportunities to connect with babies who often don't make eye contact and can exhibit repetitive behaviors and become obsessed with a single toy or other object. These and other symptoms had been identified among study subjects through the Autism Observation Scale for Infants and other clinical measures.
"These were very atypical infants, and everyone who interacted with them recognized that," Rogers said. "So I do think we found a group of children who had very atypical social communication development."
Mothers (with their babies) went through 12 weekly one-hour training and observation sessions with professional therapists. Several families received "booster" sessions or were referred to extra therapy, such as speech coaching. Babies underwent regular developmental testing, usually at intervals of three months, until they were 36 months old.
Control groups were culled from 126 infants enrolled at the MIND Institute's Infant Sibling project, which studies the development trajectory of siblings of children with autism diagnoses.
In some ways, rapid developmental progress is not surprising, particularly since mothers were teaching skills appropriate for their babies' ages, the authors acknowledge. "We're capitalizing on an incredible period of brain development and natural learning," Rogers said.
But parents of children with autism can unwittingly miss opportunities for natural learning because their infants don't transmit the expected cues – facial contact and babbling among them. Or they may opt not to disturb infants who are quietly occupied most of the time.
"If that goes on for a long time, children are having fewer and fewer chances to learn and that's going to have its effect over time," Rogers said. "And that's why it's important that children get into treatment as soon as their symptoms emerge."
Rogers cautioned that later interventions also have shown promise.
"There's no reason to think that children would do better if they're getting these interventions earlier" rather than later, she said. "And in fact, most children haven't shown their symptoms this early."
Clinical measures of autism symptoms also can be problematic.