LAUSR

Background: With the rapid spread of SARS-CoV-2 across the globe, numerous authors have noted different patient characteristics that may relate to an increased admission rate to an intensive care unit (ICU). However, little data has been presented comparing these characteristics among those who receive care at either a rural or urban emergency department (ED). Study Objective: To compare the clinical characteristics and outcomes of patients with COVID-19 admitted to the ICU from rural and urban EDs. Methods: A retrospective, multi-center cohort study of adult patients who required hospitalization between March 01, 2020 and July 01, 2020 due to confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection from two rural and one urban ED in Arizona were included in analysis. Research assistants who were blinded to the study hypothesis were trained on proper data abstraction prior to the collection of data by the study team. With adherence to a quality-controlled protocol and structured abstraction tool, research assistants manually collected patient demographics, ED laboratory values, initial vital signs, total hospitalizations, ICU admissions, and mortality in a with a one-to-one allocation ratio non-White and White patients. Comparisons of the comorbidities among patients transferred to the ICU in urban and rural hospitals were conducted with the chi-squared analysis. Factors that may predict transfer to the ICU were determined via a stepwise multivariable binomial logistic regression. Results: A total of 304 patients (175 urban and 129 rural) with confirmed SARS-CoV-2 infection were admitted to the hospital during the study period with 63 patients (24 urban vs 39 rural;OR=2.1, p=0.01) being admitted to the ICU. Of those admitted to the ICU, a total of 21 (33.3%) were female (11 urban and 10 rural). The median age of patients admitted to the ICU from the urban cohort was 66.0 years old (IQR=35.0) and from the rural cohort was 62.6 years (IQR=28).The most common comorbidity seen in both urban and rural patients admitted to the ICU was hypertension (12 [50%] urban;21 [53.8%] rural). In the overall cohort, multivariable logistic regression showed an increase in the odds of ICU admission among patients presenting with concurrent bacterial infection (p=0.043), elevated temperature (p=0.002), respiratory rate (p=0.003), white blood cells (p=0.034), and reduced hemoglobin levels (p=0.014). Across the total cohort, these factors predicted transfer to the ICU with a sensitivity of 39.5% and specificity of 95.2%. Conclusion: Patients with confirmed SARS-CoV-2 are more likely to require critical care intervention if presenting to the emergency department with concurrent bacterial infection, elevated temperature, respiratory rate, white blood cells, and reduced hemoglobin. The degree to which these factors generalize between urban and rural hospitals remains to be elucidated.