Using a pilot matched-pairs cluster-randomized control trial, we evaluated the acceptability and preliminary outcomes of universal Teacher-Child Interaction Training (TCIT-U) with students with disabilities in early special education programs. Twelve classrooms (clusters) were paired by age and type and then randomly assigned within pairs to either TCIT-U (81 students, 20 teachers) or wait-list control (63 students, 16 teachers) with services as usual. We analyzed the effects of TCIT-U on (a) teachers’ skills acquisition via masked observational coding and (b) students’ behavior and developmental functioning via teacher questionnaires. (author abstract)