Earlier sepsis antibiotic initiation may not increase overtreatment, according to a recent study.
In a multicenter mixed-methods analysis performed at four Utah hospitals, investigators examined physician-level variation in door-to-antimicrobial time—defined as the time from emergency department (ED) presentation to antimicrobial administration—and evaluated whether faster practice patterns were associated with unnecessary treatment.
The investigators evaluated 9,180 adult patients who met Sepsis-3 criteria between July 1, 2013, and January 31, 2017, and were treated by 88 attending ED physicians. The patients were 18 years or older, nontrauma, and received intravenous antimicrobials prior to ED departure. The median patient age was 63 years, 50.5% were female, and 38.6% received antimicrobial treatment more than 3 hours after arrival. On final retrospective adjudication, 8.5% (n = 778) patients were determined not to have infection at presentation. Follow-up extended to 30 days.
Physician-level variation in door-to-antimicrobial time was assessed using a linear mixed-effects model incorporating physician-level random intercepts and adjustment for patient-level covariates. After adjustment, door-to-antimicrobial times varied across the physicians. For a typical patient, the physician-estimated mean door-to-antimicrobial time was 184 minutes. The median physician-level door-to-antimicrobial time was 155 minutes, ranging from 89.5 to 218 minutes, and 5.1% of the total variation in timing was attributable to the physician. Antimicrobial timing didn't differ according to physician sex, residency training, and years of experience.
The primary analysis used a joint mixed-effects shared parameter model linking physician mean door-to-antimicrobial time with the probability of overtreatment, defined as antimicrobial administration in the ED with infection ruled out on final adjudication. For every 10-minute increase in physician mean door-to-antimicrobial time, there was no statistically significant association with overtreatment. In a secondary analysis among 9,072 patients who received antibiotics, antimicrobial spectrum scores weren't associated with physician timing patterns.
Qualitative interviews conducted between May 17, 2022, and June 28, 2023, with 18 physicians in the fastest and slowest quartiles showed differences in care processes. The physicians in the fastest quartile described proactive communication, early multidisciplinary coordination, and parallel task execution; whereas the physicians in the slowest quartile described more sequential and reactive approaches to sepsis evaluation and treatment.
Limitations included the use of quantitative data from 2013 to 2017, preceding more recent acceleration in sepsis treatment practices, and the time gap between measured practice patterns and interviews. Although models adjusted for multiple patient characteristics, residual confounding or nonlinear associations may have influenced the results. The study was conducted within one regional health system in Utah, which may have limited generalizability.
“Physician practice patterns characterized by earlier antimicrobial administration were not associated with increased antimicrobial overtreatment,” noted lead study author Ithan D. Peltan, MD, MSc, of the Department of Pulmonary and Critical Care Medicine at the Intermountain Medical Center, and colleagues.
Dr. Peltan, reported institutional research funding from Novartis AG, Bluejay Diagnostics Inc, and Regeneron. Co–study authorJorie Butler, PhD, reported consulting for University of California, San Francisco; received grant support from the Rockefeller Foundation, the Department of Veterans Affairs Health Systems Research, the National Institutes of Health (NIH), and the Agency for Healthcare Research and Quality; and received salary and office resources from the Veterans Health Administration outside the submitted work. Co–study author Angela P. Presson, PhD, reported receiving support from the National Center for Advancing Translational Sciences of the NIH. Co–study author Tom H. Greene, PhD, reported receiving research funding from National Kidney Foundation, The Collaborative Study Group, the NIH, and the Patient-Centered Outcomes Research Institute outside the submitted work. Co–study author Matthew H. Samore, MD, reported receiving grant support from the Centers for Disease Control and Prevention (CDC) and the VA outside the submitted work. Co–study author Catherine L. Hough, MD, MSc, reported receiving grant support from the NIH and CDC outside the submitted work. The study authors reported no other disclosures.
Source: JAMA Network Open
