Association between primary care practice characteristics and emergency department use in a medicaid managed care organization

Background: Many patients use emergency departments (EDs) for primary care. Previous studies have found that patient characteristics affect ED utilization. However, such studies have led to few policy changes. Objectives: We sought to determine whether Medicaid patients' ED use is associated with characteristics of their primary care practices. Research Design: This was a cohort study. Subjects: A total of 57,850 patients, assigned to 353 primary care practices affiliated with a Medicaid HMO, were included. Measures: Predictor variables were characteristics of primary care practices, which were measured by visiting each practice. The outcome variable was ED use adjusted for patient characteristics. Results: On average, patients made 0.80 ED visits/person/yr. Patients from practices with more than 12 evening hours/wk used the ED 20% less than patients from practices without evening hours. A higher ratio of the number of active patients per clinician-hour of practice time was associated with more ED use. When more Medicaid patients were in a practice, these patients used the ED more frequently. Other factors associated with ED use included equipment for the care of asthma and presence of nurse practitioners and physician assistants. Discussion: Modifiable characteristics of primary care practices were associated with ED use. Because the observational design of this study does not allow definitive conclusions about causality, future studies should include intervention trials to determine whether changing practice characteristics can reduce ED use. Conclusions: Improving primary care access and scope of services may reduce ED use. Focusing on systems issues rather than patient characteristics may be a more productive strategy to improve appropriate use of emergency medical care.