Purpose Few studies have examined the content of teaching rounds on general internal medicine ward services. Even less is known about what makes for effective attending teaching rounds with regard to resident and medical student education. We hypothesized that both students and residents would feel that rounds were educational if they were more actively involved.
Methods Over a four month period, trained research assistants (second year medical students) observed attending rounds at our university hospital. The research assistants coded the two most dominant activities occurring within each 5-minute interval during rounds with respect to content and involvement. Activities on rounds were conceptualized as involving either teaching, patient care, administrative tasks or dead time. Involvement was coded as to the participation of individual team members: attending physician, upper level resident, interns and students. For the purpose of analysis, each team member's involvement was calculated as the percentage of the total rounding time that they were judged to be involved in the discussions. After rounds, the research assistants rated the rounds with regards to teaching quality (1 = poor, 10 = excellent) and the perceived student/resident involvement (1 = not involved, 10 = very involved).
Results Overall, 96 different attending rounds were observed. The teaching quality of rounds was judged to be above average with a mean rating of 6.2 ± 2.2. Overall residents were judged to be more involved than students, 8.4 ± 1.7 versus 5.1 ± 2.6. Likewise, examining the percentage of rounds involving different team members, attending physicians were the most involved (accounting for 72% of the interactions on rounds), followed by residents (41%), interns (37%) and finally students (23%). The overall teaching quality as judged by the research assistants was associated with higher rating of perceived student involvement (r = .37, p = .0005) but not with the perception of resident involvement (r = .12, p = .24). Examining the percentage of rounding time involving different team members, research assistants judged rounds to be of higher teaching quality when the attending physicians were less involved (r = -.25, p = .02).
Conclusions The majority of teaching rounds are centered on the attending physician with medical students having the lowest involvement of all team members. However, the teaching that occurs on attending rounds is judged to be of higher quality when medical students are more involved and attending physicians are less involved.