Diagnoses of autism spectrum disorder (ASD) are reliable in children as young as 14 months, new data show. That is 2 to 3 years younger than the age at which children usually receive the diagnosis — a difference that could allow children to receive treatment during a critical growth period for the brain.
The findings underscore the need for early detection and repeat testing, the authors write. An independent expert agrees.
“[C]hildren who are evaluated earlier have more stability in their diagnosis than we previously thought (ie, the diagnosis persists), and [that] tells us that children should be referred for diagnostic evaluation and treatment as early as possible,” said Susan E. Levy, MD, MPH, vice chair, Committee for Protection of Human Subjects, Center for Autism Research, Roberts Center for Pediatric Research, Philadelphia, Pennsylvania, in an interview with Medscape Medical News.
“There are strong data that children who are referred earlier for early intervention and developmental treatment services have better outcome,” continued Levy, who was not involved in the study. “Early intervention is helpful for all children who have delays identified — it supports families, provides intervention to support future development, and tracks the children’s progress.”
The prospective cohort study, by Karen Pierce, PhD, from the Department of Neurosciences, University of California, San Diego, La Jolla, and colleagues, was published online April 29 in JAMA Pediatrics.
From 2006 through 2018, a total of 2241 toddlers underwent diagnostic evaluation at the autism expert evaluation center after either screening positive in primary care or receiving a community referral. Of those children, 1269 were eligible for the current study, having undergone their first diagnostic evaluation at age 12 to 36 months and having undergone at least one subsequent evaluation.
Of the toddlers in the study, 918 (72.3%) were boys. The median age at initial evaluation was 17.6 months, and the median age at final evaluation was 36.2 months.
Children received one of the following diagnoses: ASD (n = 441), ASD features (78), developmental delay (89), language delay (80), other issue (91), typically developing (TD) (439), or typical sibling of an ASD proband (51).
Overall, most children designated as having ASD initially had been diagnosed with ASD at their last visit. The diagnostic stability was higher for ASD than any of the other diagnoses, at 0.84 (95% confidence interval [CI], 0.80 – 0.87).
An analysis based on 2-month age bands showed that diagnostic stability for ASD was weakest at age 12 to 13 months (stability coefficient, 0.50; 95% CI, 0.32 – 0.69). Diagnostic stability rose to 0.79 by age 14 months and to 0.83 by age 16 months.
Although ASD diagnostic stability was lowest among children in the earliest age band, half of the children continued to receive that diagnosis at 3 years. Among those who did not continue to receive the diagnosis, most “ended up having some other delay or disorder,” Pierce said.
The American Academy of Pediatrics recommends that all children undergo screening with a standardized autism-specific screening tool at the 18-month preventive care visit. However, the US Preventive Services Task Force has found insufficient evidence to determine the benefits and harms of screening for ASD in children aged 18 to 30 months unless there are specific concerns regarding ASD.
Pierce says children should be screened starting at 12 months.
“Chances are you’re not even going to get your diagnostic evaluation appointment until 14 months. Even if you did get it slightly earlier, if [a child] failed the ASD-specific testing, chances are there is something wrong,” Pierce said.
For those children whose diagnosis changed from the first to the last visit, the most common transition was from an initial diagnosis of language or developmental delay to ASD. Only rarely did children transition from an initial designation of ASD to a final diagnosis of TD. Of 400 toddlers initially diagnosed as having ASD, seven transitioned to a final diagnosis of TD. Five of those were first evaluated at age 12 to 13 months.
“The diagnostic [accuracy for] autism started to get really strong at 14 months, but we started screening at 12 months,” Pierce said. She explained that there is often a waiting list for an appointment for a diagnostic evaluation. “Continue to screen at 12 months and then have parents try to get an appointment that falls somewhere closer to the 14-month point, because if you wait until 14 months, who knows when they’re going to get their evaluation?”
Continued Vigilance Needed
Almost one quarter of toddlers (23.8%) were not designated as having ASD at their initial visit but were diagnosed at a later visit. “[Twenty-four percent] of the toddlers that eventually got diagnosed with ASD when they turned 3 were sort of missed by clinicians. They failed the screening, they came in for their evaluation, and they just looked like they had a language delay or a developmental delay,” Pierce explained.
“Then when they came back for repeated evaluations, surprise — they started failing the test for autism and started meeting criteria…. So the message for clinicians is that if you see a child and they’re looking like they might just have a mild delay, it doesn’t mean they’re out of the woods; they still could meet criteria and start showing stronger signs of autism as they get slightly older, so that means you should do a repeat screening,” Pierce said.
Levy agrees with that approach. “Clinicians should continue to conduct developmental and behavioral surveillance at each well-child visit (eg, ‘Do you have any concerns about your child’s development or behavior?’), and the provider should be on the lookout for any concerns they may observe or hear about as part of any other history. Autism-specific developmental screening should be done at 18 and 24 months,” she told Medscape Medical News.
Clinicians should take any of the family’s concerns seriously, particularly in cases in which children are at an increased risk for ASD, such as those with a genetic disorder, prematurity, or an older sibling with ASD, she added.
“If either [informal or specific screening] is positive, the provider should refer the child and family to early intervention for evaluation and services and, if appropriate, to a specialist,” said Levy, who is also affiliated with the Division of Developmental and Behavioral Pediatrics, the Children’s Hospital of Philadelphia, and is a professor of pediatrics at Perelman School of Medicine at the University of Pennsylvania.
“If you’ve checked a couple of times across that 12-month period between 12 and 24 months, then I think there’s a much better chance of not missing anyone,” Pierce said.
Late Diagnosis Is a Missed Opportunity
Autism is believed to develop during the first and second trimesters of pregnancy, but research has shown that the human brain is capable of much development and remodeling, especially during the early postnatal years, the researchers explain.
“The few studies that have examined treatment during this transformative time window have found that toddlers with ASD, cerebral palsy, premature birth, and severe hearing loss experience significant positive changes, such as an increase in 15 IQ points or improvements in speech perception and language ability,” the authors write.
“My thinking is if we get in there and start treatment before the [synaptic] connections have happened, we can provide particular environmental stimulation, particular social and play therapies, that could potentially encourage more positive and effective connections to happen,” Pierce told Medscape Medical News.
“If we wait until after 3 or 4 [years], those connections are already made, so now you have the job of having to resculpt and reshape what’s already there. It seems logical that it would be much easier, and you’d have a better outcome, if you provided the environmental stimulation while the connections are being formed,” Pierce explained.
Pierce plans to follow the children identified in this study while they’re in school to determine whether this earlier diagnosis and treatment made a difference. “This is what we believe, but we have to test it scientifically and empirically,” she said.
The authors and Levy have disclosed no relevant financial relationships.
JAMA Pediatrics. Published online April 29, 2019. Full text