Findings and implications of developmental screening for high-risk children referred to a tertiary developmental disability center

A Prediction Model for Detecting Developmental Disabilities in Preschool-Age Children Through Digital Biomarker-Driven Deep Learning in Serious Games: Development Study

Background

Early detection of developmental disabilities in children is essential because early intervention can improve the prognosis of children. Meanwhile, a growing body of evidence has indicated a relationship between developmental disability and motor skill, and thus, motor skill is considered in the early diagnosis of developmental disability. However, there are challenges to assessing motor skill in the diagnosis of developmental disorder, such as a lack of specialists and time constraints, and thus it is commonly conducted through informal questions or surveys to parents.

Objective

This study sought to evaluate the possibility of using drag-and-drop data as a digital biomarker and to develop a classification model based on drag-and-drop data with which to classify children with developmental disabilities.

Methods

We collected drag-and-drop data from children with typical development and developmental disabilities from May 1, 2018, to May 1, 2020, via a mobile application (DoBrain). We used touch coordinates and extracted kinetic variables from these coordinates. A deep learning algorithm was developed to predict potential development disabilities in children. For interpretability of the model results, we identified which coordinates contributed to the classification results by applying gradient-weighted class activation mapping.

Results

Of the 370 children in the study, 223 had typical development, and 147 had developmental disabilities. In all games, the number of changes in the acceleration sign based on the direction of progress both in the x- and y-axes showed significant differences between the 2 groups (P<.001; effect size >0.5). The deep learning convolutional neural network model showed that drag-and-drop data can help diagnose developmental disabilities, with an area under the receiving operating characteristics curve of 0.817. A gradient class activation map, which can interpret the results of a deep learning model, was visualized with the game results for specific children.

Conclusions

Through the results of the deep learning model, we confirmed that drag-and-drop data can be a new digital biomarker for the diagnosis of developmental disabilities.

Associations Among Referral Concerns, Screening Results, and Diagnostic Outcomes of Young Children Assessed in a Statewide Early Autism Evaluation Network

OBJECTIVES

To examine associations between referral concerns, screening results, and diagnostic outcomes for young children evaluated across a statewide primary care network for early screening and diagnosis of autism spectrum disorder (ASD).

STUDY DESIGN

The Early Autism Evaluation Hub system was developed to increase developmental screening and improve access to timely ASD evaluations in local communities. In 2019, 858 children (ages 18-48 months; 40% diagnosed with ASD) received ASD evaluations across 12 Early Autism Evaluation Hubs. Data on primary care provider (PCP)- and caregiver-reported referral concerns, Modified Checklist for Autism in Toddlers, Revised with Follow-Up (MCHAT-R/F) and Ages and Stages Questionnaire, Third Edition (ASQ-3), and diagnostic outcome were collected.

RESULTS

Among children evaluated, there was low concordance between PCP and caregiver referral concern. Although a positive MCHAT-R/F screen was associated with PCP but not caregiver-reported ASD referral concern, there was a significant linear relationship between MCHAT-R/F raw scores and both PCP and caregiver ASD referral concern. A different pattern of ASQ-3 delays was found to be associated with PCP-reported as compared with caregiver-reported ASD referral concern. Finally, PCP-reported ASD referral concern, positive MCHAT-R/F, and ASQ-3 Communication and Personal Social delays were associated with a significantly higher likelihood of subsequent ASD diagnosis.

CONCLUSIONS

Understanding how community PCPs use surveillance and screening data, the extent to which PCPs and caregivers have shared understanding and engage in collaborative decision-making about evaluation referral, and how these factors relate to diagnostic outcomes has the potential to impact educational efforts for both PCPs and caregivers of young children, as well as inform the development of more efficacious early identification approaches.

Referral, Assessment and Use of Screening Measures Related to Autism Spectrum Disorder at a Tertiary Hospital Setting

Children with developmental concerns in Australia continue to experience inequitable healthcare and service-related delays, even when diagnostic risk is identified. This study sought to explore service and demographic pathway factors leading up to autism spectrum disorder (ASD) assessment, including value of screening measures applied at triage. Following a trial of centralised intake for referred young children with suspected ASD, observational, retrospective pathway data was explored using bivariate and regression analyses. The mean age of 159 children referred with autism symptoms was 3.6 years, and 64% were diagnosed with ASD. Service allocation was associated with diagnosis, whilst screening tool results were not. Improved pathways are needed to limit wasted waiting times and direct each child to needs-based services.

A Game-Based Repeated Assessment for Cognitive Monitoring: Initial Usability and Adherence Study in a Summer Camp Setting

The current feasibility study examined the adherence, reliability, and assessment potential of an evidence-based game-like mobile Monitoring Tool (Akili Interactive Labs), to monitor 100 participants' cognition for eight sessions at a summer camp for children with special needs. A validated measure of attention was administered at baseline. In the last session, participants completed an exit questionnaire. The Monitoring Tool was found to be enjoyable, and showed a high rate of adherence. No Monitor-related adverse events were reported. Monitor metrics showed good reliability across repeated measurements, indicating it is stable over long-term cognitive monitoring. There was evidence that the Monitoring Tool was able to detect differences in cognition between the children diagnosed with attention deficit hyperactivity disorder and autism spectrum disorders.

Comparative use of the Ages and Stages Questionnaires in the USA and Scandinavia: a systematic review

AIM

The aim of this systematic review was to investigate screening practices with the Ages and Stages Questionnaires (ASQ) and the Ages and Stages Questionnaires: Social-Emotional (ASQ:SE) in the USA and Scandinavia and to identify practical lessons and research opportunities.

METHOD

The review was performed for ASQ- and ASQ:SE-related studies in children from birth to 5 years. From nine databases and 1689 references (published from 1988-2018), 127 articles were included and categorized using Covidence online software. The Critical Appraisal Skills Programme Checklists were used before data synthesis.

RESULTS

US studies primarily use the ASQ/ASQ:SE to detect delays in general and at-risk populations in medical settings, which increases early detection, clinician-referral, and intervention rates. Scandinavian studies commonly use the ASQ/ASQ:SE to monitor developmental-behavioural differences in intervention/exposure-based cohorts. Pre-visit screening yields completion/return rates of 83% to more than 90% and fosters same-day interpretation. When referrals are indicated, systemwide care coordination or colocation with a developmental-behavioural specialist is beneficial.

INTERPRETATION

Practical implementation lessons are reviewed. Research opportunities include investigating and measuring the ASQ/ASQ:SE's 'overall' sections. Danish, Norwegian, and Swedish translations are available but up-to-date norming and validation studies are needed throughout Scandinavia. Randomized controlled trials are needed to investigate outcomes in screened versus unscreened cohorts.

WHAT THIS PAPER ADDS

General and at-risk populations broadly benefited from periodic Ages and Stages Questionnaires (ASQ) and/or Ages and Stages Questionnaires: Social-Emotional (ASQ:SE) screening. Pre-visit ASQ and/or ASQ:SE screenining implementation systems work best. The ASQ and ASQ:SE 'overall' sections are not quantifiable and under-researched.

Screening for Autism Spectrum Disorder: Profiles of Children Who Are Missed

OBJECTIVE

To characterize children presenting with concerns for autism spectrum disorder (ASD) missed by parent-report screeners and to examine benefits of a combined screening approach with the Modified Checklist for Autism in Toddlers, Revised (M-CHAT-R) and the Ages and Stages Questionnaire, Third Edition (ASQ-3).

METHODS

Participants included were 154 children aged 16 to 42 months presenting for an evaluation at an autism center. Caregivers completed the M-CHAT-R, ASQ-3, and a demographic questionnaire. Children participated in an autism diagnostic evaluation consisting of the Mullen Scales of Early Learning (Mullen) and Autism Diagnostic Observation Schedule, Second Edition (ADOS-2).

RESULTS

A total of 124 children (81%) were diagnosed with ASD. The M-CHAT-R identified 85% (n = 105) of these children. Children with ASD missed by the M-CHAT-R had significantly higher scores on the Mullen and significantly lower scores on the ADOS-2. Of the ASQ-3 domains, the majority (n = 102, 82%) of children with ASD failed the communication domain; missed cases showed similar patterns of higher Mullen scores and lower ADOS-2 scores. When adopting a combined screening approach, using a failed screen from either the M-CHAT-R or ASQ-3 communication domain, 93% of children were identified. Parent-reported concerns on an open-ended questionnaire revealed ASD red flags for many missed cases.

CONCLUSION

Children with ASD missed by screeners had higher scores on developmental testing and lower scores on the ADOS-2; however, children still performed below average on developmental tests. Our findings suggest that a combined screening approach was most effective for identifying children with ASD from a sample group referred for an ASD evaluation.