Background: In the past two decades, several screening instruments have been developed to detect toddlers who may be autistic, both in clinical and unselected samples. Among others, the Quantitative CHecklist for Autism in Toddlers (Q−CHAT) is a quantitative and normally distributed measure of autistic traits which demonstrated good psychometric properties in different settings and cultures. Recently machine learning (ML) has been applied to behavioural science to improve classification performance of autism screening and diagnostic tools, but mainly in children, adolescents and adults.
Methods: In this study, we used machine learning (ML) to investigate the accuracy and reliability of the Q−CHAT in discriminating young autistic children from those without. Three different ML algorithms (Random Forest, Naive Bayes and Support Vector Machine) were applied to investigate the complete set of Q-CHAT items and the best predictive items.
Results: Our results showed that the three selected models outperformed the classical statistical methods of predictive validity and among the three ML classifiers, the Support Vector Machine was the most effective, being able to classify autism with 95% accuracy. Furthermore, using the Support Vector Machine-Recursive Feature Elimination approach we were able to select a subset of 14 items ensuring an accuracy of 93%, while an accuracy of 83% was obtained from the best 3 discriminating items in common to our and the previous reported Q-CHAT-10.
Limitations: Further data collection is needed.
Conclusions: This evidence confirms the high performance and cross-cultural validity of the Q-CHAT and supports the application of ML to create shorter and faster versions of the instrument maintaining high classification accuracy, to be used as a quick, easy and high-performance tool in primary care settings.