Why It's Important to Continue Universal Autism Screening While Research Fully Examines Its Impact.

McPheeters et al1 provided a systematic review of the evidence thatwas used by theUSPreventive Services Task Force (USPSTF) to make its recent recommendations on universal autism screening.2 The USPSTF concluded that there is currently insufficientevidenceto assess the benefits of universal screening for autismspectrumdisorder (ASD) inyoung children. Inparticular, theUSPSTFnoted that no studyhasdirectly compared the long-term outcomes of screened vs nonscreened children and, further, studies of the efficacy of early ASD treatment have not been based on samples identified through screening. Thus, theUSPSTFconcludes that the longterm benefits and harms of screening in the general population cannot yet be determined. The report was met with considerable controversy. The president of the American Academy of Pediatrics (AAP) promptly issued a statement noting that the USPSTF recommendations“runcounter toAAPguidelines”andthat the“AAP remains committed to its recommendation for the timely screeningand identificationof childrenwhobenefit fromearly intervention and treatment.”3 Autism advocacy organizations andUSRepsChris SmithandMikeDoyle, co-chairs of the Congressional Autism Caucus, urged the USPSTF to reconsider their recommendations on universal autism screening. What then is the basis of this controversy? The AAP and other professional and advocacy organizations that supportuniversal autismscreeninghavebased their currentpracticeguidelinesonseveral research findings,which are described in the reviewbyMcPheeters et al.1 First, autism is aprevalent condition, affecting 1 in68children in theUnited States, and associated with significant shortand long-term burdens.4 The lifetime cost of supporting an individual with ASD is estimated to be $1.4 million. That cost is $2.4 million for those who also have an intellectual disability.5 Second, current autism screening tools, while not perfect, adequately detect ASD in children between approximately 18 and 36months in age.6,7 Themost commonly used tool is aparentquestionnaire that requires little trainingby the health careprofessionals and takes less than5minutes to complete. If a child’s score falls in a questionable range, then an additional 5 to 10minutes of follow-up questions are needed. Large studies conducted in community primary care settings have found that, of those children who fail the screen, about 50%will bediagnosedwithASD, and98%will have anactionable developmental concern.6,7 Although the data on false negatives are scant, the available evidence suggests that the current screening tool does not miss a substantial number of children. All children who fail the screen can be referred for infant-toddler assessment and intervention services through Part C of the Individuals with Disabilities and Education Act, a programsupported by a federal grant established in 1986by Congress. Third, studies have shown that screening leads to earlier referral and diagnosis.8 Most children identified through screeninghavenotbeenpreviously identifiedbyeither thepediatrician or a parent. Childrenwho are screened, on average, arediagnosedearlier and thus receive servicesat anearlier age. This finding is especially significant forAfricanAmerican and Hispanic children, who tend to be diagnosed and access services at a later age than do white children.9 The majority of states have passed legislation that mandates insurance benefits for early intervention for children with diagnosed ASD. Finally, studies evaluating the efficacy of early intervention showsignificant improvements in cognitive and languageoutcomes, andayounger ageof enrollment in treatmenthasbeen shown to result in better outcomes.10,11 In short, researchhas shownthatuniversal screening leads to earlier diagnosis andearlier access to intervention and, further, that earlier intervention leads to better outcomes; however, no studyhas yet been conducted that examineswhether screening,per se, leads tobetteroutcomes. Suchastudywould require a large representative group of children from the general population to be randomly assigned to receive or not receive autism screening and then be followed up longitudinally to assess the impact of screening on the long-term outcomes of these children and their families, as well as the overall societal cost. Studies on the efficacy of early intervention insamplesofchildrenyounger than3yearswhowere identified through screening are also needed. Interventions appropriate for children with ASD younger than 3 years have been developed and evaluated. A systematic review of 24 studies related to intervention for children withASDyounger than3yearswas recentlypublished.12Compared with interventions for preschool-aged children, interventions for children younger than 3 years more often intensively involve parents and use naturalistic developmental behavioral approaches.13 Such interventions usually are delivered byparents in the context of everyday routines and the parents’ interactionswith their children, and all are designed topromote learning, engagement, andcommunicationand reduce behavioral challenges. The review identified 6 randomized clinical trials that assessed the efficacy of intervention in childrenwithASDyounger than3years,with treatments rangRelated article at jama.com Opinion