Sally Ozonoff

Autism screening aids in early detection and diagnosis. Several screeners are available for toddlers 16 months and older, but there is little data on the accuracy of screening measures used in the first year of life. A community sample ( N = 1641) completed three screeners—M‐CHAT‐R, ITC, VIRSA—online at 6, 9, 12, 18, and 24 months of age (including exploratory use of M‐CHAT‐R prior to 16 months). At 36 months, outcomes of ASD or Non‐ASD were determined by tele‐evaluation and outcome surveys. Psychometric properties were assessed for all screeners at all ages, both used individually and in combination. When used in the first year of life (6–12 months), psychometric properties were in the poor range, with area under the curve (AUC) < 0.70 for all three screeners, whether used in isolation or combination. At 18 and 24 months, the M‐CHAT‐R and ITC outperformed the VIRSA, with psychometric properties in the fair range (AUC 0.70–0.79) used in isolation and the good range (AUC ≥ 0.80) when used in combination. Using the M‐CHAT‐R and ITC at more than one age in the second year also fell in the good range. This study, using procedures to address limitations of prior screening studies and follow screen‐negative cases, found that none of the three screeners had adequate psychometric properties in the first year. At 18 and 24 months, findings supported the use of multiple screeners, either in combination or longitudinally, for best prediction of autism diagnosis.

Autism screening in childhood is common, yet little is known about its potential psychological impact on caregivers. The U.S. Preventive Services Task Force, an independent national panel of disease prevention experts, stated that this gap in knowledge limited their ability to endorse universal autism screening. This study examined the psychological impact of autism screening, using data from a large community‐based sample ( n = 1272) involving online caregiver‐completed autism screeners at child age 6, 9, 12, 18, and 24 months. Caregivers completed the Participation Impact Questionnaire retrospectively (mean child age at completion 37.2 ± 4.8 months) to measure feelings about screening. A minority (34.7%) of the sample reported presence of ≥ 1 negative feeling; the most commonly endorsed was “worried”. Among this subset, negative feelings were of short duration (lasted for < 1 day in 56.9%), were mild in severity (86.4%), and did not affect functioning (85.3%). A majority (86.2%) also reported ≥ 1 positive feeling. Our findings address a critical evidence gap regarding potential harms of autism screening and support universal screening, given that psychological harms are not common and have low functional impact, as well as possible psychological benefits.

Prospective studies of later born siblings of autistic individuals often focus on predicting autism diagnosis. Studies concentrating on siblings who do not develop autism have found subclinical atypicalities in some children as early as the first year of life. However, when followed to school-age, the continuity of these findings has been mixed. We tracked nonautistic siblings (n = 151 higher familial likelihood of autism, n = 115 lower likelihood) longitudinally from infancy to 6-16 years of age when participants completed a battery of social communication measures (parent report and direct observation/administration). Using latent profile analysis, we derived groupings based on patterns of performance across measures. Three groups were identified: Class 1 (45.5%), Class 2 (45.2%), and Class 3 (9.3%)-characterized by higher, intermediate, and lower school-age social communication abilities, respectively. We then examined the performance of these classes on independent measures of pragmatic language, reciprocal social interaction, and cognition. Class 3 demonstrated social communication differences that were most evident with novel interactive partners (e.g., examiners) and scored lower on IQ and academic achievement measures, indicating that social communication differences captured by the latent profile analysis were part of a broader pattern of developmental differences. Using data collected in the first 3 years of life, we found that the school-age classes began showing differences by 12-18 months of age-evidence of continuity between early behavior and later development. Findings suggest that when early childhood challenges are observed in siblings of autistic children, even those not meeting criteria for autism, they should be monitored over time and additional support offered as needed. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

Most telehealth studies have focused on caregiver satisfaction with autism spectrum disorder (ASD) assessments. Few have examined caregiver perceptions of direct-to-home telehealth evaluations for children with developmental concerns broader than ASD. The current study explored caregiver perspectives (N = 439) of the telehealth process, including technology barriers and satisfaction with remote delivery of diagnostic feedback, before and after a 24-month tele-assessment, in children from four outcome groups (ASD, Speech-Language Concerns, Other Developmental Concerns, and No Concerns). The results indicated overall high levels of caregiver satisfaction and acceptability across all outcome groups, although the ASD group had a lower rate (89%) than the other groups (99–100%) of rating “very true” to feeling like a valued partner in the tele-assessment, the evaluation meeting their needs (89% v. 94–99%), and understanding the results (86% v. 96–98%). Families of children with Speech-Language Concerns, the largest diagnostic outcome group, were highly satisfied and found tele-assessment acceptable. There were no differences in caregiver perceptions of the tele-assessment process based on whether they had prior concerns about their child or not. Few technology barriers were reported by any groups and socio-demographic variables (race, ethnicity, socioeconomic status, travel distance from medical center) were not associated with differences in technology access or satisfaction with the tele-evaluations. Overall, these findings support the use of tele-assessment for a wide range of children whose families may not have access to in-person assessment services, as well as suggest ways to further improve remote models of care.

Objective: This study focused on in-home use of a tele-assessment tool for autism diagnosis in young children, the TELE-ASD-PEDS (TAP). Psychometric properties, caregiver experiences, and perceptions of feasibility were examined among families likely to experience barriers to in-person evaluation. Methods: One hundred eighty-two children between 18 and 42 months of age were recruited because of positive screening, primary care concerns, or early intervention referral. All participants completed initial tele-assessment including the TAP. Approximately 2 weeks later, a second visit was conducted: 92 were randomized to a repeat TAP administration by telehealth and 90 were randomized to an in-person evaluation. Caregivers completed surveys regarding challenges with technology and satisfaction with telehealth assessment. Results: Overall, 77% of the sample was diagnosed with autism spectrum disorder (n = 140). There were few diagnostic disagreements (n = 10, 6%) between initial and second evaluations, with disagreements equally distributed between second visit type. Diagnostic outcomes (autism vs no autism) agreed between telehealth and in-person evaluation for 94% of cases, kappa = 0.82. Outcomes agreed between 2 telehealth visits for 94% of cases, kappa = 0.84. Test-retest reliability of total TAP scores across 2 administrations was strong, intraclass correlation coefficient = 0.85. Very few caregivers reported challenges with technology during telehealth appointments (<6%); 92% reported that there was nothing they would change about the telehealth visit. Examiners also reported high satisfaction with telehealth assessments. Conclusion: In-home use of the TAP is valid, reliable, feasible, and acceptable to caregivers. Findings support the continued use of telehealth assessment for autism in toddlers, which can reduce disparities in access to timely diagnostic services.

Prevalence of autism diagnosis has historically differed by demographic factors. Using data from 8224 participants drawn from the Environmental influences on Child Health Outcomes (ECHO) Program, we examined relationships between demographic factors and parent‐reported autism‐related traits as captured by the Social Responsiveness Scale (SRS; T score > 65) and compared these to relations with parent‐reported clinician diagnosis of ASD, in generalized linear mixed effects regression analyses. Results suggested lower odds of autism diagnosis, but not of SRS T > 65, for non‐Hispanic Black children (adjusted odds ratio [OR] = 0.76, 95% CI 0.55, 1.06) relative to non‐Hispanic White children. Higher maternal education was associated with reduced odds of both outcomes (OR = 0.73, 95% CI 0.51, 1.05 for ASD autism diagnosis and 0.4, 95% CI 0.29, 0.55 for SRS score). In addition, results suggested a lower likelihood of autism diagnosis but a higher likelihood of an SRS score > 65 in Black girls. Findings suggest lower diagnostic recognition of autism in non‐Hispanic Black children, despite a similar degree of SRS‐assessed autism‐related traits falling in the clinically elevated range. Further work is needed to address this disparity.

Evidence suggests core autism trait consistency in older children, but development of these traits is variable in early childhood. The Social Responsiveness Scale (SRS) measures autism‐related traits and broader autism phenotype, with two age‐dependent forms in childhood (preschool, 2.5–4.5 years; school age, 4–18 years). Score consistency has been observed within forms, though reliability across forms has not been evaluated. Using data from the Environmental Influences on Child Health Outcomes (ECHO) program (n = 853), preschool, and school‐age SRS scores were collected via maternal report when children were an average of 3.0 and 5.8 years, respectively. We compared reproducibility of SRS total scores (T‐scores) and agreement above a clinically meaningful cutoff (T‐scores ≥ 60) and examined predictors of discordance in cutoff scores across forms. Participant scores across forms were similar (mean difference: 3.3 points; standard deviation: 7), though preschool scores were on average lower than school‐age scores. Most children (88%) were classified below the cutoff on both forms, and overall concordance was high (92%). However, discordance was higher in cohorts following younger siblings of autistic children (16%). Proportions of children with an autism diagnoses were also higher among those with discordant scores (27%) than among those with concordant scores (4%). Our findings indicate SRS scores are broadly reproducible across preschool and school‐age forms, particularly for capturing broader, nonclinical traits, but also suggest that greater variability of autism‐related traits in preschool‐age children may reduce reliability with later school‐age scores for those in the clinical range.

Early motor delays and differences are common among children with autism spectrum disorder (ASD) and attention‐deficit/hyperactivity disorder (ADHD). Yet, little work has shown whether there are early atypical motor signs that differentiate these groups. Quantitative measures of movement variability hold promise for improving the identification of subtle and specific differences in motor function among infants and toddlers at high likelihood for ASD and ADHD. To this end, we created a novel quantitative measure of movement variability (movement curvature) and conducted a preliminary investigation as to whether this measure improves outcome predictions. We used a wearable triaxial accelerometer to evaluate continuous motion‐based activity in infants at high and low likelihood for ASD and ADHD at 12, 18, 24, and 36 months of age. At 36 months, participants were categorized into three outcome groups: ASD (n = 19), ADHD concerns (n = 17), and a comparison group (n = 82). We examined group differences in movement curvature and whether movement curvature is predictive of a later ASD or ADHD concerns classification. We found that movement curvature was significantly lower in infants with later ASD diagnosis at 18, 24, and 36 months of age compared to infants with either ADHD concerns or those in the comparison group. Movement curvature was also a significant predictor of ASD at 18, 24, and 36 months (AUC 0.66–0.71; p = 0.005–0.039) and when adjusting for high ASD likelihood at 18 and 24 months (AUC 0.90, p = 0.05–0.019). These results indicate that lower movement curvature may be a feature of early motor differences in infants with later ASD diagnosis as early as 18 months of age.

Autism Spectrum Disorder (ASD) presents significant challenges in early diagnosis and intervention, impacting children and their families. With prevalence rates rising, there is a critical need for accessible and efficient screening tools. Leveraging machine learning (ML) techniques, in particular Temporal Action Localization (TAL), holds promise for automating ASD screening. This paper introduces a self-attention based TAL model designed to identify ASD-related behaviors in infant videos. Unlike existing methods, our approach simplifies complex modeling and emphasizes efficiency, which is essential for practical deployment in real-world scenarios. Importantly, this work underscores the importance of developing computer vision methods capable of operating in naturilistic environments with little equipment control, addressing key challenges in ASD screening. This study is the first to conduct end-to-end temporal action localization in untrimmed videos of infants with ASD, offering promising avenues for early intervention and support. We report baseline results of behavior detection using our TAL model. We achieve 70% accuracy for look face, 79% accuracy for look object, 72% for smile and 65% for vocalization.