Chesluk B, Bernabeo E, Hess BJ, Holmboe ES, Lynn LA, Reddy S. — American Board of Internal Medicine
Presented: Society of General Internal Medicine, April 2010
Background: Health care is delivered by teams, though these teams are often not recognized nor supported. As part of the American Board of Internal Medicine's increasing focus on work-based assessment—certifying physicians' competency in practice—we are researching whether, and how, to assess physicians' ability to work as part of a team.
Methods: Our interdisciplinary research group used a 4-step process: In step 1, we reviewed the literature on teamwork in health care and other fields (aviation, management, etc.). In step 2, we convened a panel of experts in teamwork assessment and improvement. These two steps enabled us to develop a conceptual model and a draft tool to assess competency in teamwork. For step 3, we conducted ethnographic research in three primary care practices and three hospitals, to observe teams in action and to understand the role of the physician. At this stage, the draft instrument was refined, and we focused our research on hospitalists and their teams. In step 4, we collected feedback on our constructs and instrument via "cognitive lab" sessions with 10 non-physician professionals (nurses, pharmacists, etc.).
Results: Our initial review and expert panel (steps 1 and 2) suggested that "teamwork" combines competencies in communicating, collaborating, boundary spanning (mitigating the harmful effects of hierarchy) and assessing the team environment (staying aware of the team's context and resources). To assess this complex competency, we created a tool that pairs self-assessment with feedback from non-physician collaborators, using a Likert-type scale to rate key behaviors (timely exchange of information, respectful communication, admitting mistakes, etc.).
Fieldwork (step 3) showed that primary-care physicians may have little contact with their teams, while hospitalists vary widely in both the amount and quality of their collaboration with teammates. In both contexts, the practice/hospital structure greatly affects physicians' ability to interface with their team. Cognitive lab sessions (step 4) helped define what other hospital professionals require from physicians, leading us to add items on timely communication and eliminate items implying that team members need to delegate tasks to, or require direct updates from, physicians.
This research process led us to define an innovative definition of "teamwork" as a competency for physicians, combining communication skills, awareness of others' roles and understanding of the team's context. Our research identified the need to provide a tool for physicians to "map" their own teams, and refined our assessment by highlighting the areas that non-physicians found most important–particularly, respectful, timely communication; knowledge of other professions' scope of practice; and inclusion of teammates in patient-care decisions. The next step will be pilot-testing the instrument to examine these constructs.
Conclusions: Teamwork is a vital competency for physicians, but few recognize that they work in interdisciplinary teams, understand the various dimensions of interdisciplinary teamwork, or have organizational support in assessing and improving teamwork. The organizational structure and culture of many primary care practices and hospitals hamper effective interdisciplinary teamwork. In general, an iterative, mixed-methods research process can lead to richer theoretical definitions and innovative approaches to assessment of complicated, multifaceted competencies such as teamwork.
For more information about this presentation, please contact Research@abim.org.