Universal screening for autism spectrum disorder in primary care is feasible, but the current screening tool noticeably underperformed, especially in children of color and in those from lower-income households, new research found.

Guidelines for universal autism spectrum disorder (ASD) screening currently conflict, Whitney Guthrie, PhD, of the Children’s Hospital of Philadelphia, and associates wrote in Pediatrics. The American Academy of Pediatrics recommends universal screening in children aged 18 months and 24 months to expedite earlier identification and diagnosis. However, the U.S. Preventive Services Task Force has concluded that “there is insufficient evidence to recommend universal screening, in part because of limited data on outcomes for children who screen negative and from diverse samples.”

Dr. Guthrie and associates conducted a study of 25,999 children aged 16-26 months who had a well-child visit between January 2011 and July 2015 at a Children’s Hospital of Philadelphia site that had implemented universal electronic screening. Of this group, 43% were white, 37% were black, and the remainder were Asian or of other/multiple races; 92% were non-Hispanic. The median parental income was $59,597, 54% had private insurance and 45.3% had public insurance/Medicaid, and 42% came from an urban primary care site while the rest came from suburban primary care.

Screening rates were good over the study period, with 91% of children undergoing at least one screen with the Modified Checklist for Autism in Toddlers with Follow-Up (M-CHAT/F). While 50% were screened more than once, only 48% of children were screened at 18 and 24 months, as per the AAP guideline. Children who were screened multiple times were more likely to be white and non-Hispanic, to be from a suburban site, and to have higher incomes and private insurance.

After the first M-CHAT/F screen, 9.5% of children were positive, a rate comparable with that seen in other large-scale, U.S.-based studies. Of the 2,256 children who tested positive, 41% received a second screen; 782 (95%) of these children tested negative.

After the study period, most children (n = 20,437; 87%) continued receiving care with the Children’s Hospital of Philadelphia system and had diagnostic data available past the age of 4 years. ASD prevalence was 2%, giving the M-CHAT/F a sensitivity of 39%, a specificity of 95%, a positive predictive value of 15%, and a negative predictive value of 99%.

M-CHAT/F sensitivity was higher in older children (49% at 21-26 months vs. 35% at 16-20 months) and with repeated screenings (40% vs. 32%), and positive predictive value was lower in girls (8% vs. 20%). Specificity and positive predictive value were higher in white children (98% and 24%, respectively), compared with black children (92% and 12%, respectively), Asian children (90% and 11%, respectively), and those from other/multiple racial groups (94% and 13%, respectively). Higher-income families also saw increased specificity (97% vs. 92%) and positive predictive value (20% vs. 12%), compared with lower-income families.

While Dr. Guthrie and associates wrote of new methods of screening, such as parental reporting tools supported by picture or video and “direct data-gathering methods that leverage technological advances in computing and machine learning,” Lonnie Zwaigenbaum, MD, MSc, and Jonathon Maguire, MD, MSc, argued in an editorial, also published in Pediatrics, that the M-CHAT/F “remains a strong candidate” for universal ASD screening, despite the notable weaknesses.

“Ultimately, the potential added value of ASD screening must be considered relative to what would occur in its absence,” wrote Dr. Zwaigenbaum of the Women’s and Children’s Health Research Institute at the University of Alberta, Edmonton, and Dr. Maguire of St Michael’s Hospital, Toronto, and the University of Toronto. “Although it is difficult to object to the guidance from the U.S. Preventive Services Task Force to listen carefully to parents’ concerns, we must acknowledge the false dichotomy between screening and surveillance in this context. ... Why not use the best available measurement tools to identify developmental concerns with the highest possible accuracy?”

The study was funded by the Allerton Foundation, Eagles Charitable Foundation, and the National Institute of Mental Health; the study investigators reported that they had no conflicts of interest. Dr. Zwaigenbaum is a member of an independent data monitoring committee for a Roche-funded medication trial and Dr. Maguire reported receiving nonfinancial support from Ddrops for an investigator-initiated study on vitamin D and respiratory tract infections. Dr Zwaigenbaum is supported by the Stollery Children’s Hospital Foundation Chair in Autism Research. Dr Maguire is supported by St. Michael’s Hospital and the Hospital for Sick Children.

SOURCEs: Guthrie W et al. Pediatrics. 2019 Sep 27. doi: 10.1542/peds.2018-3963; Zwaigenbaum L, Maguire J. Pediatrics. 2019 Sep 27. doi: 10.1542/peds.2019-0925.