Objective Estimates of Direct-Medical Costs Among Persons Aged 3 to 38 Years With and Without Research-Defined Autism Spectrum Disorder Ascertained During Childhood: A Population-Based Birth-Cohort Study

Gene×environment interactions in autism spectrum disorders

There is credible evidence that environmental factors influence individual risk and/or severity of autism spectrum disorders (hereafter referred to as autism). While it is likely that environmental chemicals contribute to the etiology of autism via multiple mechanisms, identifying specific environmental factors that confer risk for autism and understanding how they contribute to the etiology of autism has been challenging, in part because the influence of environmental chemicals likely varies depending on the genetic substrate of the exposed individual. Current research efforts are focused on elucidating the mechanisms by which environmental chemicals interact with autism genetic susceptibilities to adversely impact neurodevelopment. The goal is to not only generate insights regarding the pathophysiology of autism, but also inform the development of screening platforms to identify specific environmental factors and gene×environment (G×E) interactions that modify autism risk. Data from such studies are needed to support development of intervention strategies for mitigating the burden of this neurodevelopmental condition on individuals, their families and society. In this review, we discuss environmental chemicals identified as putative autism risk factors and proposed mechanisms by which G×E interactions influence autism risk and/or severity using polychlorinated biphenyls (PCBs) as an example.

A machine learning-based diagnostic model for children with autism spectrum disorders complicated with intellectual disability

Background

Early detection of children with autism spectrum disorder (ASD) and comorbid intellectual disability (ID) can help in individualized intervention. Appropriate assessment and diagnostic tools are lacking in primary care. This study aims to explore the applicability of machine learning (ML) methods in diagnosing ASD comorbid ID compared with traditional regression models.

Method

From January 2017 to December 2021, 241 children with ASD, with an average age of 6.41 ± 1.96, diagnosed in the Developmental Behavior Department of the Children's Hospital Affiliated with the Medical College of Zhejiang University were included in the analysis. This study trained the traditional diagnostic models of Logistic regression (LR), Support Vector Machine (SVM), and two ensemble learning algorithms [Random Forest (RF) and XGBoost]. Socio-demographic and behavioral observation data were used to distinguish whether autistic children had combined ID. The hyperparameters adjustment uses grid search and 10-fold validation. The Boruta method is used to select variables. The model's performance was evaluated using discrimination, calibration, and decision curve analysis (DCA).

Result

Among 241 autistic children, 98 (40.66%) were ASD comorbid ID. The four diagnostic models can better distinguish whether autistic children are complicated with ID, and the accuracy of SVM is the highest (0.836); SVM and XGBoost have better accuracy (0.800, 0.838); LR has the best sensitivity (0.939), followed by SVM (0.952). Regarding specificity, SVM, RF, and XGBoost performed significantly higher than LR (0.355). The AUC of ML (SVM, 0.835 [95% CI: 0.747-0.944]; RF, 0.829 [95% CI: 0.738-0.920]; XGBoost, 0.845 [95% CI: 0.734-0.937]) is not different from traditional LR (0.858 [95% CI: 0.770-0.944]). Only SVM observed a good calibration degree. Regarding DCA, LR, and SVM have higher benefits in a wider threshold range.

Conclusion

Compared to the traditional regression model, ML model based on socio-demographic and behavioral observation data, especially SVM, has a better ability to distinguish whether autistic children are combined with ID.

The Autism Intervention Research Network on Physical Health (AIR-P) Research Agenda

OBJECTIVES

In the United States, autistic individuals experience disproportionate physical and mental health challenges relative to non-autistic individuals, including higher rates of co-occurring and chronic conditions and lower physical, social, and psychological health-related quality of life. The Autism Intervention Research Network on Physical Health (AIR-P) is an interdisciplinary, multicenter research network for scientific collaboration and infrastructure that aims to increase the life expectancy and quality of life for autistic individuals, with a focus on underserved or vulnerable populations. The current paper describes the development of the AIR-P Research Agenda.

METHODS

Development of the research agenda involved an iterative and collaborative process between the AIR-P Advisory Board, Steering Committee, and Autistic Researcher Review Board. The methodology consisted of 3 phases: (1) ideation and design, (2) literature review and synthesis; and (3) network engagement.

RESULTS

Six core research priorities related to the health of autistic individuals were identified: (1) primary care services and quality, (2) community-based lifestyle interventions, (3) health systems and services, (4) gender, sexuality, and reproductive health, (5) neurology, and (6) genetics. Specific topics within each of these priorities were identified. Four cross-cutting research priorities were also identified: (1) neurodiversity-oriented care, (2) facilitating developmental transitions, (3) methodologically rigorous intervention studies, and (4) addressing health disparities.

CONCLUSIONS

The AIR-P Research Agenda represents an important step forward for enacting large-scale health-promotion efforts for autistic individuals across the lifespan. This agenda will catalyze autism research in historically underrepresented topic areas while adopting a neurodiversity-oriented approach to health-promotion.

IQ in Autism Spectrum Disorder: A Population-Based Birth Cohort Study

OBJECTIVES

We aimed to describe the intellectual ability and ratio of boys to girls with average or higher IQ within autism spectrum disorder (ASD) cases identified in a population-based birth cohort. We hypothesized that research-identified individuals with ASD would be more likely to have average or higher IQ, compared to clinically diagnosed ASD. We also hypothesized the male to female ratio would decrease as the definition of ASD broadened.

METHODS

ASD incident cases were identified from 31 220 subjects in a population-based birth cohort. Research-defined autism spectrum disorder, inclusive criteria (ASD-RI) was based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision, autistic disorder (AD), Asperger Disorder, and pervasive developmental disorder not otherwise specified criteria. Research-defined autism spectrum disorder, narrow criteria (ASD-RN) was a narrower definition based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision AD criteria. Clinical diagnoses of ASD were abstracted from medical and school records. Intellectual ability was based on the last IQ score or on documented diagnoses of intellectual disability if no scores available. Average or higher IQ was defined as IQ ≥86.

RESULTS

A total of 59.1% of those with ASD-RI (n = 890), 51.2% of those with ASD-RN (n = 453), and 42.8% of those with clinically diagnosed autism spectrum disorder (n = 187) had average or higher IQ. Within the ASD-RI and ASD-RN groups, boys were more likely than girls to have an average or higher IQ (62.0% vs 51.3% [P = .004] and 54.1% vs. 42.5% [P = .03], respectively).

CONCLUSION

Our data suggest that nearly half of individuals with ASD have average or higher IQ. Boys with ASD are more likely to have average or higher IQ than girls. Patients with ASD and higher IQ remain at risk for not being identified.

Ensemble classification of autism spectrum disorder using structural magnetic resonance imaging features

Background

Autism spectrum disorder (ASD) is characterized by a spectrum of social and communication impairments and rigid and stereotyped behaviors that have a neurodevelopmental origin. Although many imaging studies have reported structural and functional alterations in multiple brain regions, clinically useful diagnostic imaging biomarkers for ASD remain unavailable.

Methods

In this study, we applied machine learning (ML) models to regional volumetric and cortical thickness data from the largest structural magnetic resonance imaging (sMRI) dataset available from the Enhancing Neuro Imaging Genetics Through Meta-Analysis (ENIGMA) consortium (1833 subjects with ASD and 1838 without ASD; age range: 1.5-64; average age: 15.6; male/female ratio: 4.2:1).

Results

The highest classification accuracy on a hold-out test set was achieved using a stacked Extra Tree Classifier. The area under the receiver operating characteristic (ROC) curve (AUC) was 0.62 (95% confidence interval [CI]: 0.57, 0.68) and the area under the precision-recall curve was 0.58. Learning curve analysis showed the good fit of the model and suggests that more training examples will not likely benefit model performance.

Conclusions

Our results suggest that sMRI volumetric and cortical thickness data alone may not provide clinically sufficient useful diagnostic biomarkers for ASD. Developing clinically useful imaging classifiers for ASD will benefit from combining other data modalities or feature types, such as functional MRI data and raw images that can leverage other machine learning (ML) techniques such as convolutional neural networks.

Polychlorinated Biphenyls (PCBs): Risk Factors for Autism Spectrum Disorder?

Autism spectrum disorder (ASD) includes a group of multifactorial neurodevelopmental disorders defined clinically by core deficits in social reciprocity and communication, restrictive interests and repetitive behaviors. ASD affects one in 54 children in the United States, one in 89 children in Europe, and one in 277 children in Asia, with an estimated worldwide prevalence of 1-2%. While there is increasing consensus that ASD results from complex gene x environment interactions, the identity of specific environmental risk factors and the mechanisms by which environmental and genetic factors interact to determine individual risk remain critical gaps in our understanding of ASD etiology. Polychlorinated biphenyls (PCBs) are ubiquitous environmental contaminants that have been linked to altered neurodevelopment in humans. Preclinical studies demonstrate that PCBs modulate signaling pathways implicated in ASD and phenocopy the effects of ASD risk genes on critical morphometric determinants of neuronal connectivity, such as dendritic arborization. Here, we review human and experimental evidence identifying PCBs as potential risk factors for ASD and discuss the potential for PCBs to influence not only core symptoms of ASD, but also comorbidities commonly associated with ASD, via effects on the central and peripheral nervous systems, and/or peripheral target tissues, using bladder dysfunction as an example. We also discuss critical data gaps in the literature implicating PCBs as ASD risk factors. Unlike genetic factors, which are currently irreversible, environmental factors are modifiable risks. Therefore, data confirming PCBs as risk factors for ASD may suggest rational approaches for the primary prevention of ASD in genetically susceptible individuals.