Classroom observations have long been considered a necessary component of Attention-Deficit/Hyperactivity Disorder (ADHD) evaluations. Yet, research evaluating the utility of observational ratings in ADHD assessment is limited. This study examined the contributions of the Behavioral Observation of Students in Schools (BOSS) to ADHD assessment by investigating associations between BOSS scores with ADHD symptom clusters and symptoms of frequently co-occurring externalizing and internalizing disorders. The utility of BOSS scores in predicting future ADHD-related impairment beyond standard parent and teacher ratings was also examined. One hundred and thirty-five children in grades 2-5 across 23 public schools participated in a randomized controlled trial examining a psychosocial treatment for ADHD. BOSS ratings were collected at baseline. Parent and teacher ratings of child symptoms and impairment were collected at baseline, post-treatment (3-4 months later), and follow-up (8-12 months later). Multiple regressions investigated the associations between the BOSS subscale of Task Engagement (TE) and parent and teacher ratings of ADHD symptoms and related disorders. Multi-level modeling accounted for school cluster effects. Results showed that lower BOSS TE was related to higher teacher-rated inattention but not hyperactivity/impulsivity symptoms. Lower BOSS TE was also associated with higher teacher-rated Oppositional Defiant Disorder (ODD) and depression symptoms, but not anxiety symptoms. Further, BOSS TE predicted higher future impairment beyond baseline teacher and parent ratings of ADHD symptoms and impairment, controlling for treatment. The BOSS appears sensitive to symptoms of child inattention, ODD, and depression, and may have utility in informing future impairment beyond standard informant ratings of ADHD.