Team functioning as a predictor of patient outcomes in early medical home implementation

BACKGROUND: New models of patient-centered primary care such as the patient-centered medical home (PCMH) depend on high levels of interdisciplinary primary care team functioning to achieve improved outcomes. A few studies have qualitatively assessed barriers and facilitators to optimal team functioning; however, we know of no prior study that assesses PCMH team functioning in relationship to patient health outcomes. PURPOSE: The aim of the study was to assess the relationships between primary care team functioning, patients' use of acute care, and mortality. METHODOLOGY/APPROACH: Retrospective longitudinal cohort analysis of patient outcomes measured at two time points (2012 and 2013) after PCMH implementation began in Veterans Health Administration practices. Multilevel models examined practice-level measures of team functioning in relationship to patient outcomes (all-cause and ambulatory care-sensitive condition-related hospitalizations, emergency department visits, and mortality). We controlled for practice-level factors likely to affect team functioning, including leadership support, provider and staff burnout, and staffing sufficiency, as well as for individual patient characteristics. We also tested the model among a subgroup of vulnerable patients (homeless, mentally ill, or with dementia). RESULTS: In adjusted analyses, higher team functioning was associated with lower mortality (OR = 0.92, p = .04) among all patients and with fewer all-cause admissions (incidence rate ratio [IRR] = 0.90, p < 0.01), ambulatory care-sensitive condition-related admissions (IRR = 0.91, p = .04), and emergency department visits (IRR = 0.91, p = .03) in the vulnerable patient subgroup. CONCLUSION: These early findings give support for the importance of team functioning within PCMH models for achieving improved patient outcomes. PRACTICE IMPLICATIONS: A focus on team functioning is important especially in the early implementation of team-based primary care models.