Factors Associated with Optimizing Empiric Antibiotic Therapy in Six Hospitals
Methods: : To describe factors associated with adjustments to inpatient antimicrobial therapy we created a retrospective cohort of 1,200 adult inpatients over 6 US hospitals, hospitalized >72 hours, on 1 or more antimicrobial. Infectious disease specialists reviewed charts for comorbidities, antimicrobial therapy, imaging studies, culture results. The spectrum of antimicrobial therapy was ranked from narrow to broad, and regimens were categorized as unchanged, narrowed or discontinued, and escalated by the 5th day of antimicrobial use.
Results: : Across the 6 study sites, 60.4% (4,119/6,812) of inpatients received antimicrobials. Of 1,200 randomly selected patients with active antimicrobials, 730 (60.8%) received nonprophylactic empiric therapy and had a length of stay >3 days and were included in the analysis. At the start of therapy, 30.1% of patients were afebrile and had a normal white blood cell count. Appropriate cultures were collected from 59.2% of patients, and a little over half of those were negative (table). By the 5th day of therapy, 12.5% of empiric antimicrobials were escalated, 21.5% were narrowed or discontinued, and 66.4% were unchanged. In a multivariable generalized linear and latent mixed model, narrowing or discontinuation by day 5 was more likely to occur when cultures were collected at the start of therapy (adjusted odds ratio (OR) 1.58, 95% confidence interval (CI) 1.04–2.4) and no infection was noted on an initial radiological study (OR 1.75, 95% CI 1.18–2.59). Escalation was associated with multiple infection sites (OR 2.65, 95% CI 1.4–5.02) and the presence of a positive culture (OR 1.97, 95% CI 1.19–3.25).
Conclusion: Across a sample of US hospitals, broad-spectrum antimicrobial therapy was common, even when clinical signs of infection were absent. Fewer than 1 in 3 inpatients on empiric antimicrobials have their regimens narrowed within 5 days of starting, even at hospitals with formal antimicrobial stewardship programs.
M. Schweizer, None
D. Z. Uslan, None
T. Kelesidis, None
S. A. Weisenberg, None
B. Johannsson, None
H. Young, None
J. Cantey, None
E. Perencevich, None
E. Septimus, sage and molnlyke: received product, provided product for ABATE study
AHRQ, CDC, NIH: Grant Investigator, Grant recipient
R. Laxminarayan, None