To investigate (1) the quantity and quality of verbal feedback provided to residents in surgery and anesthesiology in the operating room (OR), (2) the relationship between provided verbal feedback and assigned entrustment scores; and (3) residents' perceptions of usefulness of verbal feedback they received in the OR for a specific entrustable professional activity (EPA). This was a prospective observational study. An academic tertiary care hospital in Kingston, Ontario, Canada. Twenty residents (8 anesthesiology, 8 general surgery, 4 orthopedic surgery) and 23 attending physicians (11 anesthesiology, 9 general surgery, 3 orthopedic surgery) from Kingston Health Sciences Centre participated voluntarily. Participants were recruited using convenience sampling and no compensation was provided. We recorded 1180 verbal feedback events across 122 hours of OR interactions for 50 attending physician-resident dyads. Verbal feedback events focused predominantly on analyzing performance (57%) and fostering learner agency (20%). Verbal feedback was rated by residents as "extremely useful" or "very useful" over 70% of the time. Verbal feedback events that addressed a greater number of FQI domains were perceived to be more useful by residents. There were no significant correlations between verbal feedback event quantity or quality and entrustment scores on EPAs. Attending physicians frequently provide residents with verbal feedback on their performance in the OR; however, the quality of this feedback is variable. Future research should focus on developing strategies to capture and utilize verbal feedback as assessment for learning within CBME.