Last week, findings from a small but notable University of California–Davis study showed that intervening earlier than usual with infants who show signs of autism spectrum disorder (ASD) can help prevent them from developing the disorder by the time they reach age 3 or 4. For perspective, we spoke with Einstein’s Dr. Lisa Shulman, a neurodevelopmental pediatrician and featured contributor to this blog. She explains how the study’s findings mirror what she sees in clinical practice at Einstein’s Children’s Evaluation and Rehabilitation Center and offers takeaways for clinicians and parents.
What are the most important findings of this study?
A small group of seven infants, 7 to 15 months of age, who were already showing early signs of autism spectrum disorder received a behavioral intervention. The intervention involved a parent-coaching model in which parents were trained by experienced clinicians in 12 weekly hour-long sessions with a curriculum derived from an evidence-based intervention program called the Early Start Denver Model, developed for older toddlers with ASD by the first author of the U.C. Davis study.
At age 3, six of the seven children did not meet the criteria for ASD and did not demonstrate the developmental delays often associated with ASD. The majority of these infants were being observed at this young age because an older sibling has ASD and they are at increased risk.
If this small pilot study is replicated, the implication is that if ASD symptoms are identified and intervention is undertaken extremely early by parents who have been taught various skills designed to increase infant engagement, an ASD diagnosis can possibly be avoided in some cases.
You’ve been diagnosing ASD for more than 20 years; how do the findings in this study correspond to what you see clinically?
Every year, we see a number of infants ages 6 to 15 months referred because of concerns about autism or other developmental delays. While some children have clear biological abnormalities or risk factors presumed to cause their patterns of delayed development, or highly atypical patterns of development, there is always a group of young infants and toddlers who have signs that provoke concern but whose social engagement can be worked on through various techniques. We talk about their having the building blocks of social capability, but they lack social motivation or contend with sensory sensitivities that get in the way of their social engagement.
This latter group can be wooed into social interaction in a variety of ways. Parents can be taught to read their children’s cues and develop their infants’ nascent social interest. We speak to parents about their children “hanging in the balance,” with the potential to go one way, toward ASD, or to be “lassoed back” into a pattern of social engagement.
Young children learn through their social interests and “nosiness” and their tendency to imitate. If an intervention can increase a baby’s social interest and nosiness, it can alter the entire course of how he or she learns in the important toddler years and beyond. The capabilities targeted in the Infant Start program of intervention get at the core weaknesses of young children with ASD. The strategies the program teaches parents allow them to woo these infants into engaging, and shift the balance toward social engagement. Based on my experience, the program feels clinically very much on the money.
What are some of the earliest signs that an infant may have ASD?
The symptoms that give rise to concern about ASD in infants (6 months or older) include:
- Unusual visual fixation (e.g., looking at a ceiling fan for long periods of time as a preferred activity)
- Unusual, repetitive patterns of object exploration (e.g., spending long periods of time spinning, tapping, or opening and closing things)
- Lack of communicative acts such as gestures (e.g., no hands being raised to be picked up by 9 months, no pointing by 12 months)
- Lack of sound and consonant production
- Lack of coordinated eye contact, smiling and reciprocal interaction during play with a parent
- Decreased eye contact and interest in social interaction
- Lack of response to name
- Reluctance to smile
- Lack of a tendency to imitate sounds or actions
- Tendency for a toddler to entertain himself or herself for long periods of time
How much clinical evidence exists to support clearance of ASD behaviors over time when children are diagnosed at 7 to 15 months, as they were in this study?
This study is unique because it took symptomatic infants and offered them the intervention. I am not aware of any other studies providing an intervention to such young infants with signs of ASD. In the literature on toddlers considered to have been diagnosed early, by age 2 or 2-1/2, several studies report that in this population about 15 percent of these children no longer warranted an ASD diagnosis at school age.
Our own clinical findings follow this pattern as well. There has been a literature of “optimal outcome”; for instance, there was a recent New York Times Magazine article, a cover story, discussing a study of kids who outgrew autism. The study involved a minority of children who had been diagnosed and received an intensive intervention by the ages of 2 or 2-1/2—which was considered extremely early until the U.C. Davis study.
There’s been a lot of positive attention paid to this study. What are some of its limitations? What are the barriers to implementing Infant Start?
1. The study was of a small group (seven infants).
2. The study was not randomized. The parents who chose to participate (and their children) may differ significantly from those who did not in some way that led to the outcome, rather than the results being attributable to the intervention itself.
3. It is not clear to me how easy it would be to generalize the protocol, in that not all parents can carry out a protocol such as this due to lack of capability, motivation or feasibility.
4. As the authors note, one of the key issues is how to identify such young children so they can receive the intervention. The majority of infants in this sample were referred because they were already in an “infant sib” study, having an older sibling with ASD. Generally, children are not screened for ASD by pediatricians until 18 months. How can we systematically identify children less than 12 months old with symptoms of ASD? It took these researchers three years to get seven babies who met the criteria and whose parents agreed to participate.
5. It is important to note that there were groups of infants with ASD symptoms who were excluded from the study: those with genetic syndromes, perinatal complications, prematurity or vision, hearing or motor impairments. The researchers were right to exclude these children with more-severe biological risk factors from the pilot study, as their deficits may be more “hardwired” and less amenable to intervention. These groups should , however, be included in future studies of the Infant Start model to better understand its generalizability to all infants showing signs of ASD. If there are groups less responsive to the intervention due to their risk status or deficit profile, that will need to be acknowledged to families and the media. Not all infants identified with ASD in infancy who receive evidence-based interventions may be able to have a meaningful reduction in symptoms.
Given those limitations, what are the practical takeaways from this study?
For clinicians: We have to look at how very young children with signs of ASD come to our attention. Right now no one “screens” (looks for signs of ASD) in a population about which no one has concerns. I think it will be important to disseminate the data supporting the novel idea that an infant can show signs of ASD even earlier than the age at which screening currently occurs (15 months or older), and what those signs are. If those signs are identified or concerns are expressed by parents or others who spend time with the infant, there is a lot at stake in terms of referring that infant for additional input and intervention. Then there will be the logistical issues of how to handle the intervention. Who will pay? Who will diagnose or deem a child in need of such service? Will the parents be able and motivated to be trained to implement a program of intervention? What if they can’t, or won’t?
For parents: The U.C. Davis study drives home the importance of knowing the early signs suggestive of ASD and seeking out appropriate expertise and input. Parents need to follow their instincts and not stop until their concerns are heard. Their infants’ and families’ futures may depend on it. That said, I am concerned that this way of thinking has the potential to create a generation of highly anxious parents. I am also concerned that early identification of ASD and developmental delays and the earliest of interventions will not result in avoidance of ASD or developmental delays in all children. Some children, despite the earliest identification and the best interventions, may not avoid a diagnosis of ASD or developmental delays because they have too many biological abnormalities. For example, at this time, there is no parent intervention that can eliminate the intellectual disability associated with Down syndrome, and yet the diagnosis of Down syndrome and its implications of developmental delay are evident at birth. Early intervention can help each child reach his or her potential—but that may not be a “cure” in all circumstances.