Relationships within inpatient physician housestaff teams and their association with hospitalized patient outcomes

IMPORTANCE: Improving inpatient care delivery has historically focused on improving individual components of the system. Applying the complexity science framework to clinical systems highlights the important role of relationships among providers in influencing system function and clinical outcomes. OBJECTIVE: To understand whether inpatient medical physician teams can be differentiated based on the relationships among team members, and whether these relationships are associated with patient outcomes, including length of stay (LOS), unnecessary length of stay (ULOS), and complication rates. DESIGN: Eleven inpatient medicine teams were observed daily during attending rounds for 2- to 4-week periods from September 2008 through June 2011. Detailed field notes were taken regarding patient care activities, team behaviors, and patient characteristics and outcomes. Behaviors were categorized using the Lanham relationship framework, giving each team a relationship score. We used factor analysis to assess the pattern of relationship characteristics and assessed the association between relationship characteristics and patient outcomes. SETTING: Observations occurred at the Audie L. Murphy Veterans Affairs Hospital and University Hospital in San Antonio, Texas. PARTICIPANTS: Physicians were chosen based on rotation schedules, experience, and time of year. Patients were included based on their admission to the inpatient medicine teams that were being observed. MAIN MEASURES: Relationship scores were based on the presence or absence of 7 relationship characteristics. LOS, ULOS, and complication rates were assessed based on team discussions and chart review. The association between relationships and outcomes was assessed using the Kruskal-Wallis rank sum test. RESULTS: We observed 11 teams over 352.9 hours, observing 1941 discussions of 576 individual patients. Teams exhibited a range of 0 to 7 relationship characteristics. Relationship scores were significantly associated with complication rates, and presence of trust and mindfulness among teams was significantly associated with ULOS and complication rates. CONCLUSIONS: Our findings are an important step in understanding the impact of relationships on the outcomes of hospitalized medical patients. This understanding could expand the scope of interventions to improve hospital care to include not only process improvement but also relationships among providers.