Validation of Social Responsiveness Scale for Korean Preschool Children With Autism

Performance of the Social Responsiveness Scale-2 for the Assessment of Autistic Behaviors in a Sample of Canadian Preschool-Aged Children

Behavioral traits of autism spectrum disorder (ASD) typically present in early childhood, underscoring the importance of screening tools for the early identification of ASD. The current study compared scores on the Social Responsiveness Scale-Second Edition (SRS-2) Preschool Form between the US standardization sample (n = 247) and a Canadian cohort of preschool-aged children (n = 595) recruited from the Maternal-Infant Research on Environmental Chemicals (MIREC) study. In the MIREC sample, we examined whether ASD-like traits are correlated with sociodemographic characteristics and child intellectual abilities, and how maternal ratings of social skills assessed by the SRS-2 are associated with maternal ratings of general problem behaviors. Mean total SRS-2 raw score was significantly lower in the MIREC sample (mean = 29.7, SD = 15.8) compared to the US standardization sample (mean = 41.9, SD = 26.0). Total raw score in the US standardization sample did not significantly differ between males (mean = 40.6, SD = 23.1) and females (mean = 42.8, SD = 28.7), whereas in the MIREC sample the total raw score was significantly higher among males (mean = 33.0, SD = 17.1) than females (mean = 26.6, SD = 13.9). A significantly larger proportion of the MIREC sample was White, younger in age, and had more educated parents compared to the US standardization sample. ASD-like traits were correlated with lower intellectual abilities, a less enriched home environment, more behavioral problems, and poorer adaptive skills. SRS-2 Preschool Form scores were significantly lower in the Canadian sample compared to the US standardization sample, which may reflect demographic differences between the two groups. Girls may be under-identified when SRS-2 Preschool Form norms are used for screening ASD.

Examining the Relationship Between Anxiety Severity and Autism-Related Challenges During Cognitive Behavioral Therapy for Children with Autism

PURPOSE

Using data from a randomized clinical trial evaluating cognitive behavioral therapy (CBT) for children with autism and co-occurring anxiety, this study examined the relationship between autism features and anxiety symptoms throughout CBT.

METHODS

Two multilevel mediation analyses were run which examined the mediating role of changes in anxiety for changes in two core features of autism, (a) repetitive and restrictive behaviors (RRBs) and (b) social communication/interaction impairments, between pre- and post-treatment.

RESULTS

Indirect effects between time and autism characteristics were significant for both models, indicating that as anxiety changes, so do RRBs and social communication/interaction as the outcomes respectively.

CONCLUSION

Findings suggest a bidirectional relationship between anxiety and autism features. Implications of these findings are discussed.

Different autism measures targeting different severity levels in children with autism spectrum disorder

The Childhood Autism Rating Scale™, Second Edition (CARS™-2) and Social Responsiveness Scale™, Second Edition (SRS™-2) are two measures for identifying autism symptoms. The CARS™-2 has two versions: Standard (CARS-ST) and High-Functioning (CARS-HF). To better understand their properties, this study aimed to investigate: (1) the associations among the CARS-ST, CARS-HF and the SRS™-2, and (2) the severity consistency between the CARS-ST and the CARS-HF. A sample of 125 children with autism spectrum disorder was recruited (mean age: 80.98 months, SD = 16.08). Based on Verbal Comprehension Index (VCI), children were divided into two groups: low severity level of autism spectrum disorder (LSL-ASD: VCI ≥ 80) and high severity level of autism spectrum disorder (HSL-ASD: VCI < 80). All children were evaluated with the CARS-ST and the SRS™-2, and the HF group, with the CARS-HF as well. In the LSL group, the CARS-ST and the CARS-HF had high correlation (r = 0.852, p < .001). Both versions had small to moderate correlations with the SRS™-2 (r = 0.130-0.491). In the HSL group, no significant correlations were found between the CARS-ST and SRS™-2 (p > .05). The CARS-HF and the CARS-ST had low severity consistency (Kappa = 0.376, p < .01). The CARS-ST and the CARS-HF had high correlations but low severity consistency. Different correlation patterns were found between the CARS™-2 and the SRS™-2 in the LSL and HSL groups. The results should help clinicians better understand the properties of the measures and choose appropriate measures when assessing autism symptoms.

Reliability and validity of the Autism Screen for Kids and Youth

LAY ABSTRACT

It is important that autistic children be diagnosed as early as possible so their needs can be met and their families can gain important insights into their behavior and interact with them appropriately. However, very few autism screening instruments are appropriate for children who have outgrown early childhood. The Autism Screen for Kids and Youth (ASKY) presents parents of children aged 4-18 years with 30 items that relate to autistic behaviors as defined by the current clinical diagnostic criteria for autism spectrum disorder (DSM-5 ASD). We evaluated the Hebrew instrument's performance on 167 autistic and non-autistic children and adolescents. We found that the ASKY algorithm correctly identified 92% of the autistic individuals as "probable ASD" and correctly identified 72% of the non-autistic individuals as "probable non-ASD," with these classifications showing excellent stability over time. Using total questionnaire score instead of the algorithm improved the ASKY's ability to correctly identify autistic individuals as "probable ASD" and non-autistic individuals as "probable non-ASD" to 93% and 78%, respectively. Overall, the ASKY is a promising instrument for ASD screening of older children.

Identifying major predictors for parenting stress in a caregiver of autism spectrum disorder using machine learning models

Introduction

Previous studies have investigated predictive factors for parenting stress in caregivers of autism spectrum disorder (ASD) patients using traditional statistical approaches, but their study settings and results were inconsistent. Herein, this study aimed to identify major predictors for parenting stress in this population by developing explainable machine learning models.

Methods

Study participants were collected from the Department of Child and Adolescent Psychiatry, Severance Hospital, Yonsei University College of Medicine, Seoul, the Republic of Korea between March 2016 and October 2020. A total of 36 model features were used, which include subscales of the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) for caregivers' psychopathology, Social Responsiveness Scale-2 for core symptoms, and Child Behavior Checklist (CBCL) for behavioral problems. Machine learning classifiers [eXtreme Gradient Boosting (XGBoost), random forest (RF), logistic regression, and support vector machine (SVM) classifier] were generated to predict severe total parenting stress and its subscales (parental distress, parent-child dysfunctional interaction, and difficult child). Model performance was assessed by area under the receiver operating curve (AUC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. We utilized the SHapley Additive exPlanations tree explainer to investigate major predictors.

Results

A total of 496 participants were included [mean age of ASD patients 6.39 (SD 2.24); 413 men (83.3%)]. The best-performing models achieved an AUC of 0.831 (RF model; 95% CI 0.740-0.910) for parental distress, 0.814 (SVM model; 95% CI 0.720-0.896) for parent-child dysfunctional interaction, 0.813 (RF model; 95% CI 0.724-0.891) for difficult child, and 0.862 (RF model; 95% CI 0.783-0.930) for total parenting stress on the test set. For the total parenting stress, ASD patients' aggressive behavior and anxious/depressed, and caregivers' depression, social introversion, and psychasthenia were the top 5 leading predictors.

Conclusion

By using explainable machine learning models (XGBoost and RF), we investigated major predictors for each subscale of the parenting stress index in caregivers of ASD patients. Identified predictors for parenting stress in this population might help alert clinicians whether a caregiver is at a high risk of experiencing severe parenting stress and if so, providing timely interventions, which could eventually improve the treatment outcome for ASD patients.

Psychometric properties of the Mandarin version of the autism diagnostic observation Schedule-Generic

BACKGROUND/PURPOSE

The diagnosis of autism spectrum disorder (ASD), involving multiple components of clinical assessments, is challenging. The Autism Diagnostic Observation Schedule-Generic (ADOS-G), one of the standardized and validated instruments for ASD diagnostic evaluation, has been widely used in many countries. With the preparation of the Mandarin version of the ADOS-G (Mandarin-ADOS-G), this study aims to examine its psychometric properties, including reliability and validity.

METHODS

The sample included 554 individuals clinically diagnosed with ASD (477 males, 86.1%) and 50 typically developing (TD) individuals (29 males, 58.0%) who were assessed with different modules of the Mandarin-ADOS-G between 4.1 and 34.0 years old with a mean age of 13.0 years (Module 1, n = 40; Module 2, n = 46; Module 3, n = 275; Module 4, n = 243). We evaluated the inter-rater reliability, test-retest reliability, internal consistency, and concurrent validity with the Chinese Autism Diagnostic Interview-Revised (ADI-R) and Social Responsiveness Scale (SRS) caregiver-report and self-report forms. The discriminative validity of Mandarin-ADOS-G was also examined.

RESULTS

The Mandarin-ADOS-G demonstrated good inter-rater reliability (agreement of ADOS classification 0.91), good test-retest reliability (intraclass correlations 0.55-0.73), and low to high good internal consistency (Cronbach's alpha 0.27-0.86). The concurrent validity showed significant correlations with ADI-R (Pearson correlations 0.22-0.37) and the SRS caregiver-report form (Pearson correlations 0.15-0.23). Moreover, all Mandarin-ADOS-G domains successfully differentiated autistic individuals from TD individuals (all p-values <0.001).

CONCLUSION

The Mandarin-ADOS-G is a reliable and valid instrument for assisting the diagnosis of ASD in the Mandarin-speaking population.

Machine learning based on eye-tracking data to identify Autism Spectrum Disorder: A systematic review and meta-analysis

BACKGROUND

Machine learning has been widely used to identify Autism Spectrum Disorder (ASD) based on eye-tracking, but its accuracy is uncertain. We aimed to summarize the available evidence on the performances of machine learning algorithms in classifying ASD and typically developing (TD) individuals based on eye-tracking data.

METHODS

We searched Medline, Embase, Web of Science, Scopus, Cochrane Library, IEEE Xplore Digital Library, Wan Fang Database, China National Knowledge Infrastructure, Chinese BioMedical Literature Database, VIP Database for Chinese Technical Periodicals, from database inception to December 24, 2021. Studies using machine learning methods to classify ASD and TD individuals based on eye-tracking technologies were included. We extracted the data on study population, model performances, algorithms of machine learning, and paradigms of eye-tracking. This study is registered with PROSPERO, CRD42022296037.

RESULTS

261 articles were identified, of which 24 studies with sample sizes ranging from 28 to 141 were included (n = 1396 individuals). Machine learning based on eye-tracking yielded the pooled classified accuracy of 81 % (I² = 73 %), specificity of 79 % (I² = 61 %), and sensitivity of 84 % (I² = 61 %) in classifying ASD and TD individuals. In subgroup analysis, the accuracy was 88 % (95 % CI: 85-91 %), 79 % (95 % CI: 72-84 %), 71 % (95 % CI: 59-91 %) for preschool-aged, school-aged, and adolescent-adult group. Eye-tracking stimuli and machine learning algorithms varied widely across studies, with social, static, and active stimuli and Support Vector Machine and Random Forest most commonly reported. Regarding the model performance evaluation, 15 studies reported their final results on validation datasets, four based on testing datasets, and five did not report whether they used validation datasets. Most studies failed to report the information on eye-tracking hardware and the implementation process.

CONCLUSION

Using eye-tracking data, machine learning has shown potential in identifying ASD individuals with high accuracy, especially in preschool-aged children. However, the heterogeneity between studies, the absence of test set-based performance evaluations, the small sample size, and the non-standardized implementation of eye-tracking might deteriorate the reliability of results. Further well-designed and well-executed studies with comprehensive and transparent reporting are needed to determine the optimal eye-tracking paradigms and machine learning algorithms.

Diagnostic validity of Autism Diagnostic Observation Schedule, second edition (K-ADOS-2) in the Korean population

Background

Although the Korean version of the Autism Diagnostic Observation Schedule-2 (K-ADOS‐2) is widely being used to diagnose autism spectrum disorder (ASD) in South Korea, no previous study has examined the validity and reliability of all modules of K-ADOS-2 across a wide age range, particularly older children, adolescents, and adults.

Method

Data from 2,158 participants were included (mean age = 79.7 months; 73.6% male): 1473 participants with ASD and 685 participants without ASD (Toddler Module, n = 289; Module 1, n = 642; Module 2 n = 574; Module 3 n = 411; Module 4, n = 242). Participants completed a battery of tests, including the K-ADOS or K-ADOS-2 and other existing diagnostic instruments. Sensitivity, specificity, area under the receiver operating characteristic (ROC) curve, positive predictive value (PPV), negative predictive value (NPV), Cohen’s kappa (k), and agreement with existing diagnostic instruments were computed. Cronbach’s α values were also calculated.

Results

All developmental cells of the K-ADOS-2 showed sufficient ranges of sensitivity 85.4–100.0%; specificity, 80.4–96.8%; area under the ROC curve, .90-.97; PPV, 77.8–99.3%; NPV, 80.6–100.0%; and k values, .83–.92. The kappa agreements of developmental cells with existing diagnostic instruments ranged from .20 to .90. Cronbach’s α values ranged from .82 to .91 across all developmental cells.

Limitation

The best-estimate clinical diagnoses made in this study were not independent of the K-ADOS-2 scores. Some modules did not include balanced numbers of participants in terms of gender and diagnostic status.

Conclusion

The K-ADOS-2 is a valid and reliable instrument in diagnosing ASD in South Korea. Future studies exploring the effectiveness of the K-ADOS-2 in capturing restricted, repetitive behaviors and differentiating ASD from other developmental disabilities are needed.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13229-022-00506-5.