Share this article on:

Panel Concludes Evidence Is Lacking to Support Routine Autism Screening in Young Children: US Preventive Services Task Force Issues “I” Statement; Experts Weigh in

Hiscott, Rebecca

doi: 10.1097/01.NT.0000471695.33637.15
Back to Top | Article Outline


A draft recommendation statement from the US Preventive Services Task Force states that “the current evidence is insufficient to assess the balance of benefits and harms of screening for autism spectrum disorder (ASD) in children for whom no concerns of ASD have been raised by their parents of clinical provider.” Clinicians discuss the statement and its potential implications for practice.

Figure. No caption a...

Should children younger than three be screened for autism spectrum disorder (ASD) if they are asymptomatic? Not necessarily, according to the US Preventive Services Task Force (USPSTF), which issued a draft recommendation statement in August suggesting that there is not enough available evidence to assess the benefits and harms of routine screening or early intervention for so-called “asymptomatic” children under three years of age.

“We're quite clear that we're not encouraging clinicians who have decided to screen to stop,” said David Grossman, MD, MPH, a vice-chairman for the USPSTF, emphasizing that the statement does not advocate for any change in clinical practice. “But we're also not telling clinicians who aren't screening that it's time to get going. We're telling them that they need to exercise their best clinical judgment. This is a call for research, and we itemize exactly what the gaps are.” [For more information, see “USPSTF Draft Recommendation Statement: Key Points.”]

The task force based its recommendation on a review of published studies. “In the end, there were a few areas in the literature that were deficient,” said Dr. Grossman, who is also senior investigator at the Group Health Research Institute, medical director for population health strategy and a practicing pediatrician at the Group Health health care system in Seattle, and a professor of health services and an adjunct professor of pediatrics at the University of Washington Schools of Public Health and Medicine.

For example, he cited a lack of randomized controlled trials — the evidentiary gold standard — of routine autism screening. “We don't have screening trials where we randomly assign people to get screened or not, and then look at their long-term outcomes,” he said.

Treatment trials in this population were also lacking, he said. “The [existing] trials largely focused on populations of more severely affected and older children, not the types of children you would necessarily find during routine screening.”

Children identified through routine screening — rather than after a parent expresses a concern or a clinician notes a symptom — generally have a more mild presentation, he said. These children are often not the targets of such clinical trials.

“We would encourage randomized controlled trials of therapies for children who have been identified through screening: kids with milder symptoms, who are much younger,” he said.

Dr. Grossman also emphasized that “screening is not case finding. Case finding is where parents do express concern about abnormal signs, or clinicians notice them during developmental surveillance or examinations. By all means, we would expect clinicians to act on those concerns or those signs. This is about using a screening tool in well-child care for every child that comes in the door.”

Clinicians who spoke with Neurology Today agreed that there needs to be more investigation of routine screening and the impact of earlier identification and treatment in this younger, asymptomatic population. But some felt the task force had overlooked existing evidence showing that routine screening does help identify children with milder presentations of autism, which leads to earlier intervention — which can, in some cases, lead to substantial improvements in social, behavioral, and cognitive outcomes.

“I'm worried that the outcome of this paper will be that people say, ‘Why bother?’ and stop screening,” said Susan E. Levy, MD, MPH, a pediatrician specializing in early identification and treatment of ASD at the Children's Hospital of Philadelphia, a chairperson of the Autism Subcommittee of the Council on Children with Disabilities for the American Academy of Pediatrics (AAP), and a co-author of the AAP's guidelines for autism screening in young children. [See the sidebar, “Medical Society Recommendations on Screening for Autism.”]

“I'm hoping we can educate people enough to convince them that while the evidence from this report is important, there are clear benefits to screening,” she said.

Back to Top | Article Outline


There's no question that more research needs to be done, particularly to look at the long-term outcomes of early detection, intervention, and treatment among young children with ASD and other developmental disorders who are identified through routine screening, experts told Neurology Today.

But they said the task force's rigorous methodology and emphasis on randomized controlled trials missed the breadth of evidence, both from published studies and physicians' clinical experience, which suggests that routine screening leads to earlier identification and thus earlier intervention in children with autism — which can be quite beneficial for some children.

“The points the task force raise are quite valid,” said Fred R. Volkmar, MD, the Irving B. Harris professor of child psychiatry, pediatrics, and psychology in the Child Study Center at Yale University School of Medicine, and a co-author on the AACAP autism screening recommendations.

“At the same time, the downside of not screening is that you don't pick up kids who can potentially improve dramatically with treatment. As someone who deals with thousands of kids, there's no question that we're seeing much better outcomes than we used to [with earlier intervention]. Kids with autism are going to college for the first time, for example.”

It's true that randomized controlled trials of long-term outcomes of earlier identification and intervention are lacking, he said. “We'd like to have randomized controlled trials, but sometimes that does not exist.” In that case, he said, “you have to talk about what the evidence shows, and, in the worst case — which isn't always so bad — we use clinical judgment.”

“The same intervention doesn't help all kids,” said Max Wiznitzer, MD, FAAN, a pediatric neurologist in the Neurologic Institute at University Hospitals' Rainbow Babies and Children's Hospital and a professor of pediatrics and neurology at Case Western Reserve University School of Medicine, “but we've seen the further regression of skills the longer you wait to intervene, and we've seen that intervention at an earlier point in time to some degree prevented this further loss of skills.

“The data is clearly supportive that if you intervene, kids do better than if you don't intervene,” he said.

Roberto F. Tuchman, MD, FAAN, FAAP, founding director of Autism and Neurodevelopment Programs at the Miami Children's Hospital Dan Marino Center and an associate professor of neurology at Miami Children's Hospital and the University of Miami School of Medicine, argued that early screening with a validated tool like M-CHAT is important to identify any potential problems in a child's social or cognitive function or development — even if screening does not result in an autism diagnosis.

Also, the problem with only screening “symptomatic” children, said Nancy J. Minshew, MD, FAAN, a professor of psychiatry and neurology at the University of Pittsburgh School of Medicine and director of the National Institutes of Health (NIH) Center of Excellence in Autism research at the University of Pittsburgh, is that “pediatricians and parents often do not recognize or attach significance to early manifestations of ASD. Thus, relying on parents and pediatricians to define a child as symptomatic or not symptomatic before doing a screening is a very faulty premise.”

She added that “the scientific evidence of efficacy of existing interventions for ASD is limited because funding of such studies has been limited, not because treatment does not work. Such studies are also by definition longitudinal and results evolve over time.”

Dr. Grossman acknowledged that the USPSTF holds evidence to a higher standard than do many professional society guideline statements. “Some use grading methodologies that provide for a ‘weak recommendation,’” he said. “Some might view the evidentiary process of the task force as conservative. And though we recognize that the harms of screening and treatment are likely low in this specific instance, our process also requires moderate to high certainty of benefit in the population being screened.”

He said he believes that it would be possible to conduct randomized controlled trials of autism screening in asymptomatic young children, and that the trials that exist so far “are promising.” But right now, he said, “the evidence just isn't all there.”

Ultimately, the experts who spoke with Neurology Today stressed that the draft recommendation statement does not take a firm position on the issue of screening and should not change clinical practice.

“I think that clinicians will continue to follow the best practices set forth by their professions,” said Ashura W. Buckley, MD, a pediatric neurologist with the National Institute of Mental Health at the NIH, who is part of an NIH/AAN task force that is in the process of writing recommendations on interventions for children with ASD. “I would encourage pediatricians to follow the recommendations of the AAP and to stay up to date with where the research takes us.”

Back to Top | Article Outline


Patient Population

This recommendation applies to asymptomatic children [less than 18 months of age] who have not been diagnosed with autism spectrum disorder (ASD) or developmental delay and for whom no concerns of ASD have been raised by their parents or clinical provider.

Back to Top | Article Outline


The USPSTF found adequate evidence that currently available screening tests [such as the Modified Checklist for Autism in Toddlers, or M-CHAT, and its subsequent revisions] can detect ASD in children ages 18 to 30 months.

Back to Top | Article Outline

Benefits of Early Detection and Intervention/Treatment

The USPSTF found inadequate direct evidence on the benefits of screening for ASD in toddlers and preschool-age children. There are no studies meeting inclusion criteria of clinical outcomes of children identified with ASD through screening.

The USPSTF also found inadequate evidence on the efficacy of treatment [including behavioral, medical, educational, allied health, and complementary and alternative medicine approaches] of cases of ASD detected through screening. Treatment studies were generally very small, few were randomized trials, most included children who were older than would be identified through screening, and all were in clinically-referred rather than screen-detected patients.

Back to Top | Article Outline

Harms of Early Detection and Intervention/Treatment

The USPSTF found that the harms of screening for ASD and subsequent interventions are small based on evidence about the prevalence, accuracy of screening, and likelihood of minimal harms from behavioral interventions. [Harms may include the time, effort, and anxiety associated with further testing after a positive screening result, particularly if confirmatory testing is delayed because of resource limitations.]

Back to Top | Article Outline

USPSTF Assessment

The USPSTF concludes that there is insufficient evidence to assess the benefits of screening for ASD in children. Evidence is lacking for the population to be screened, and the balance of benefits and harms cannot be determined.

Source: US Preventive Services Task Force.

Back to Top | Article Outline


Other professional society guidelines, including one co-authored by the AAN, recommend universal screening for ASD in young children, the task force statement noted. The American Academy of Pediatrics Bright Futures™ guidelines, for example, recommend continuous developmental surveillance at every well-child visit, and use of a validated screening tool such as the M-CHAT or Autism Screening Questionnaire (ASQ) at 18 and 24 months. The American Academy of Child and Adolescent Psychiatry (AACAP) recommends routine developmental assessment of young children, including questions about symptoms that could be indicative of ASD.

The AAN and the Child Neurology Society, in their guidelines, recommend that developmental surveillance be performed at all well-child visits from infancy through school age, and at any time afterwards if the parent or clinician raises concerns about a child's learning, social functioning, or behavior. If surveillance suggests a possible developmental or intellectual delay, the child should be screened with a validated instrument such as the M-CHAT or ASQ, according to the guidelines.

Back to Top | Article Outline


•. US Preventive Services Task Force: Draft Recommendation Statement: Autism Spectrum Disorder in Young Children: Screening:
•. American Academy of Pediatrics, Council on Children With Disabilities, Section on Developmental and Behavioral Pediatrics, et al. Identifying infants and young children with developmental disorders in the medical home: an algorithm for developmental surveillance and screening Pediatrics 2006;118:405–420
•. Volkmar F, Siegel M, Woodbury-Smith M, et al. Practice Parameter for the Assessment and Treatment of Children and Adolescents with Autism Spectrum Disorder J Am Acad Child Adolesc Psychiatry 2014;53(2):237–257
•. Filipek PA, Accardo PJ, Ashwal S, et al. Practice parameter: Screening and diagnosis of autism: Report of the Quality Standards Subcommittee of the American Academy of Neurology and the Child Neurology Society Neurology 2000;55(4):468–479
© 2015 American Academy of Neurology