Autism screening is critical for the most vulnerable children

Early screening is crucial for identifying children with atypical development and initiating appropriate evidence-based interventions, particularly in high-risk populations. Hamner et al. present data from a cross-sectional study of nearly 10 000 toddlers who received autism screening using the Modified Checklist for Autism in Toddlers, Revised, with Follow-Up (M-CHAT-R/F) at well-child visits.1 Results indicated elevated screen-positive and autism spectrum disorder (ASD) diagnosis rates in children born preterm, with particularly dramatic increases in children born extremely preterm (<28 weeks) and very preterm (28–32 weeks). This study adds substantial value by identifying both the elevated screening and diagnostic rates and by demonstrating very little gain when adjusting for gestational age – an issue of considerable debate for the last two decades. The fact that screen-positive individuals who did not meet criteria for ASD tended to have one or more other developmental concerns meriting additional evaluation and intervention emphasizes the need for early screening (without gestational age adjustment) in this population. This observation also highlights the need for greater attention, and ideally a medical home for children born extremely and very preterm, where clinicians can carefully evaluate at key timepoints the child's challenges and needs. These study findings also raise two other important issues for future investigation. The first issue concerns what etiological factors may drive this substantial increase in ASD and other developmental problems in those who are born extremely or very preterm. The authors mention several possibilities, but this area clearly merits more attention as molecular or neurophysiological findings may point toward potential biological intervention strategies that can mitigate or lessen the impact of pathological processes.2 The second issue concerns the value of preterm gestational age as a useful risk factor in an evidence-based assessment process. The present data suggest substantial value in altering the diagnostic probability for ASD or, more generally, for any developmental concern. However, large prospective studies will be needed that more carefully follow screen-negative cases and that work to avoid selective sampling issues that can arise with follow-up. These studies will be crucial for estimating likelihood ratios for use in an evidence-based assessment framework with other informative variables, including biological sex, family history, parent-report of autism symptoms, clinician observation, and newly-established biomarkers.3-5 Evaluation of whether autism screening (using the M-CHAT-R/F or other screeners) is sufficient for maximizing sensitivity (without massively inflating false positives) to other developmental concerns is also an important future direction. It may be that additional non-autism developmental screeners are necessary to provide adequate coverage of other developmental concerns when attempting to fully capture the population in need of further evaluation and monitoring. Thus, Hamner et al.'s study is an important step toward establishing preterm birth as an ASD risk factor, providing important guidance as to how this risk factor may be used to better identify autism and other developmental disability cases. This work should also motivate researchers to conduct both large-scale prospective studies that attempt to replicate and extend the present findings, while establishing more precise and generalizable diagnostic efficiency statistics to support evidence-based screening and evaluation, as well as detailed human and animal model studies of the effects of preterm birth on the developing brain and potential neuroprotective interventions. Not required.