• The current study explored subtypes of autism based on engagement in challenging behaviors and investigated differences in treatment response across the subgroups.
• Seven clusters were found and each cluster was characterized by a single dominant challenging behavior.
• Treatment response significantly differed for some clusters, with clusters characterized by self-injurious behavior and aggression demonstrating lower skill acquisition.
• Interventions targeting self-injurious behavior and aggression may be worth prioritizing in treatment.
Gardner-Hoag, J., Novack, M. N., Parlett-Pelleriti, C., Stevens, E., Dixon, D. R., & Linstead, E. (2021). Unsupervised machine learning for identifying challenging behavior profiles to explore cluster-based treatment efficacy in children with autism spectrum disorder: Retrospective data analysis study. JMIR Medical Informatics, 9(6), 1-16. http://dx.doi.org/10.2196/27793