Behavior analytic interventions for children with autism: Policy and practice in the United Kingdom and China

Machine learning determination of applied behavioral analysis treatment plan type

BACKGROUND

Applied behavioral analysis (ABA) is regarded as the gold standard treatment for autism spectrum disorder (ASD) and has the potential to improve outcomes for patients with ASD. It can be delivered at different intensities, which are classified as comprehensive or focused treatment approaches. Comprehensive ABA targets multiple developmental domains and involves 20-40 h/week of treatment. Focused ABA targets individual behaviors and typically involves 10-20 h/week of treatment. Determining the appropriate treatment intensity involves patient assessment by trained therapists, however, the final determination is highly subjective and lacks a standardized approach. In our study, we examined the ability of a machine learning (ML) prediction model to classify which treatment intensity would be most suited individually for patients with ASD who are undergoing ABA treatment.

METHODS

Retrospective data from 359 patients diagnosed with ASD were analyzed and included in the training and testing of an ML model for predicting comprehensive or focused treatment for individuals undergoing ABA treatment. Data inputs included demographics, schooling, behavior, skills, and patient goals. A gradient-boosted tree ensemble method, XGBoost, was used to develop the prediction model, which was then compared against a standard of care comparator encompassing features specified by the Behavior Analyst Certification Board treatment guidelines. Prediction model performance was assessed via area under the receiver-operating characteristic curve (AUROC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).

RESULTS

The prediction model achieved excellent performance for classifying patients in the comprehensive versus focused treatment groups (AUROC: 0.895; 95% CI 0.811-0.962) and outperformed the standard of care comparator (AUROC 0.767; 95% CI 0.629-0.891). The prediction model also achieved sensitivity of 0.789, specificity of 0.808, PPV of 0.6, and NPV of 0.913. Out of 71 patients whose data were employed to test the prediction model, only 14 misclassifications occurred. A majority of misclassifications (n = 10) indicated comprehensive ABA treatment for patients that had focused ABA treatment as the ground truth, therefore still providing a therapeutic benefit. The three most important features contributing to the model's predictions were bathing ability, age, and hours per week of past ABA treatment.

CONCLUSION

This research demonstrates that the ML prediction model performs well to classify appropriate ABA treatment plan intensity using readily available patient data. This may aid with standardizing the process for determining appropriate ABA treatments, which can facilitate initiation of the most appropriate treatment intensity for patients with ASD and improve resource allocation.

Parenting Stress and Parenting Efficacy of Parents Having Children with Disabilities in China: The Role of Social Support

Raising children with disabilities is challenging for parents, who experience high parenting stress. The study aimed to understand the status quo of parenting efficacy of parents having children with disabilities and to analyze the association between parent efficacy, parenting stress, and social support in China. We surveyed 373 parents having children with disabilities enrolled in special education schools or rehabilitation institutions from 14 provinces in China. The online questionnaire including Parental Stress Index-Short Form, the Child Adjustment and Parent Efficacy Scale-Developmental Disability (CAPES-DD), and Social Support Scale was applied in the study. The results suggested that parenting efficacy of parents having children with disabilities in China was at the medium level. Parenting stress, social support, and parenting efficacy were significantly related with each other, and social support played a mediating role between parenting stress and parenting efficacy. The findings indicated that reducing parenting stress and improving social support might improve parenting efficacy. We also discussed the implications of providing intervention strategies or social support to improve parenting efficacy for parents with disabled children in China.

Assessing the Satisfaction and Acceptability of an Online Parent Coaching Intervention: A Mixed-Methods Approach

Background

Parent-mediated intervention (PMI) has been studied in promoting skill acquisition or behavior change in the children with autism spectrum disorder. Most studies emphasize on the improvement of child's core symptoms or maladaptive behaviors, making parental perceived competence and self-efficacy secondary. Yet, the evaluations of intervention implementation are under-reported, especially when translating such interventions into a new population or context. This research investigated the intervention implementation of a 12-week parent coaching intervention which was delivered through telehealth and tailored to Chinese population. The intervention was based on the Parent-mediated Early Start Denver Model with culturally adapted lectures, manuals, and demonstration and commentary videos. This study aimed to evaluate the intervention implementation by assessing parents' satisfaction, acceptability, appropriateness, and feasibility.

Method

A randomized controlled trial was conducted with two telehealth conditions: self-directed and web+group therapy. Parents in the self-directed condition received intervention individually through the online learning platform. The web+group therapy condition navigated the same program with weekly 1.5-h group coaching sessions via videoconferencing. This mixed-methods study used a concurrent convergent design to evaluate the intervention implementation at post-intervention. The quantitative data was collected from the Program Evaluation Survey and the qualitative data was collected from five focus groups.

Results

Parents in self-directed group reported significantly lower scores in total perceived competence than parents in web+group therapy condition, while there was no group difference on the total self-efficacy. Tailored feedback, demonstration and commentary videos, peer commenting, live coaching, and guided reflection were the top-five acceptable telehealth strategies that were strongly endorsed by parents. Family centered care, home-based intervention, strategies relative to daily activities, the remote learning platform, and the program-based community were elements that parents considered when evaluating the program's appropriateness. Parent modeling, step-by-step instructions, and tailored feedback were key components in making intervention strategies feasible for parents to implement at home.

Conclusion

Findings indicate the application of telehealth was acceptable, appropriate, and feasible for Chinese parents. Group-based parent coaching intervention via videoconferencing could be a promising home-based service model to increase parental perceived competence. A large-scale RCT is needed to investigate the effectiveness of group-based PMI via telehealth.

Controlled Trial of a Short-term Intensive Parent Training Program within the Context of Routine Services for Autistic Children in China

Recent systematic reviews found limited rigorous research conducted to date of the effectiveness of parent training programs in reducing behavioral problems for autistic children in low- and middle-income countries. This study is aimed at evaluating the effectiveness of a short-term intensive parent training program for autistic children aged three to six in the context of routine service provision in China. A quasi-experiment was conducted involving the local implementing organization and using a waitlist control. Data were collected at baseline and immediate post-intervention. The primary outcome was child behavioral problems measured using the Child Behavior Checklist Externalizing scale. Between-group comparisons used a difference-in-differences design with propensity score weighting to reduce sources of bias. A process evaluation was undertaken in parallel to assess participant involvement, program acceptability, and delivery. The protocol was prospectively registered with ClinicalTrials.gov (NCT04257331). The final sample size was 111 (treatment: 63; comparison: 48). Results suggest that the program was associated with improvements in child externalizing behaviors (b = -2.71, 95% CI [-5.23, -0.18]), parental mental health symptoms (b = -5.96, 95% CI [-11.74, -0.17]), over-reactive parenting (b = -0.63, 95% CI [-0.98, -0.27]), and parental knowledge (b = 2.08, 95% CI [2.07, 2.17]). Exploratory analysis of factors related to implementation indicated that baseline parental mental health was related to participant engagement, and that satisfaction and engagement levels were potentially linked to positive treatment effects. Findings suggest that short-term intensive parent training programs that are provided by trained non-specialists, could potentially be used as an alternative to traditional prohibitively costly services that are delivered intensively for consecutive years in low-resource contexts. Follow-ups are needed to investigate its long-term benefits.