Background:

Interdisclinlinary rounds (IDR) are commonly used to improve collaboration among hospital team members. To our knowledge, no studies have evaluated teamwork during IDR.

Methods:

This study was conducted on six general medicine units (three teaching and three nonteaching hospitalist units) at an urban 897–bed hospital. All units conducted structured interdisciplinary rounds (SIDR), which combines a structured format for communication with a forum for regular interdisciplinary meetings. SIDR occurred each weekday in unit conference rooms and lasted 30–40 min. The nurse manager and unit medical director co–led SIDR, which was attended by all nurses and physicians on the unit, as well as the pharmacist, social worker, and case manager for the unit. We adapted the Observational Teamwork Assessment for Surgery (OTAS) tool, a behaviorally anchored rating scale shown to be reliable and valid in surgical settings. The OTAS tool provided scores ranging from 0 to 6 (0 = problematic behavior; 6 = exemplary behavior) across five domains (communication, coordination, cooperation/backup behavior, leadership, and monitoring) and for prespecified subteams, which we defined as physicians, nurses, social work–case management (SW–CM), pharmacy, and coleaders. Two researchers conducted unannounced direct observations of SIDRs in October–November, 2011. We used intraclass correlation coefficients (ICCs) to assess inter–rater reliabiilty for 10 joint observations. We used Kruskal–Wallis tests to assess for differences across units, domains, and subteams. We used Spearman rank correlation to assess for correlation between time spent in discussion per patient and overall unit score.

Results:

We conducted 7–8 independent observations for each unit (total = 44). SIDR lasted 41.4 [pm] 11.1 min and 98.1% of patients were discussed (1.5 [pm] 0.38 min/patient). Inter–rater reliability was high at the unit level (ICC = 0.88) and across domains (ICC = 0.76–0.84) but was variable across subteams (ICC > 0.7 for pharmacy and coleaders; ICC = 0.64 for SW–CM; ICC <= 0.50 for physicians and nurses). We found significant differences in teamwork scores across units, with a median (IQR) 4.38 (3.92–4.88) for the lowest and 5.44 (5.28–5.52) for the highest performing unit (p < 0.01). Domain scores also differed significantly with leadership receiving the lowest score [median (IQR) = 5.0 (4.6–5.3)] and cooperation and monitoring receiving highest scores [median (IQR) = 5.4 (5.0–5.6) and 5.4 (5.0–5.7)]. Subteam scores ranged from a median (IQR) 5.0 (4.4–5.8) for coleaders to 5.5 (5.0–5.8) for SW–CM but differences were of borderline signficance (p = 0.05). We found no relationship between unit teamwork score and time spent in discussion per patient.

Conclusions:

Variation in teamwork performance during SIDR was evident across units and domains. Providing performance feedback to units may complement efforts to improve teamwork during SIDR.