Rural Patients with Severe Sepsis or Septic Shock Who Bypass Rural Hospitals Have Increased Mortality: An Instrumental Variables Approach

Objective: To identify factors associated with rural sepsis patients' bypassing rural emergency departments to seek emergency care in larger hospitals, and to measure the association between rural hospital bypass and sepsis survival. Design: Observational cohort study. Setting: Emergency departments of a rural Midwestern state. Patients: All adults treated with severe sepsis or septic shock between 2005 and 2014, using administrative claims data. Interventions: Patients bypassing local rural hospitals to seek care in larger hospitals. Measurements and Main Results: A total of 13,461 patients were included, and only 5.4% (n = 731) bypassed a rural hospital for their emergency department care. Patients who initially chose a top-decile sepsis volume hospital were younger (64.7 vs 72.7 yr; p < 0.001) and were more likely to have commercial insurance (19.6% vs 10.6%; p < 0.001) than those who were seen initially at a local rural hospital. They were also more likely to have significant medical comorbidities, such as liver failure (9.9% vs 4.2%; p < 0.001), metastatic cancer (5.9% vs 3.2%; p < 0.001), and diabetes with complications (25.2% vs 21.6%; p = 0.024). Using an instrumental variables approach, rural hospital bypass was associated with a 5.6% increase (95% CI, 2.2-8.9%) in mortality. Conclusions: Most rural patients with sepsis seek care in local emergency departments, but demographic and disease-oriented factors are associated with rural hospital bypass. Rural hospital bypass is independently associated with increased mortality.