877. Declines in Broad-spectrum Antimicrobial Prescribing for Children Hospitalized with Community-Acquired Pneumonia following Publication of a National Guideline
Session: Poster Abstract Session: Pediatric Antimicrobial Stewardship
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C

Background: Published in August 2011, the PIDS/IDSA community-acquired pneumonia (CAP) guideline recommends narrow-spectrum antimicrobials for most children hospitalized with CAP. We assessed the impact of this guideline on antimicrobial prescribing among children enrolled in the CDC Etiology of Pneumonia in the Community (EPIC) Study between Jan 2010-Jun 2012.

Methods: Children hospitalized with CAP were prospectively enrolled at two children's hospitals in TN and one in UT (N=2628). Treatment decisions, including antibiotic selection, were made independent of study protocol. The monthly proportions of children exclusively receiving 3rd generation cephalosporins with or without a macrolide (broad-spectrum) or aminopenicillins with or without a macrolide (narrow-spectrum) during the first 2 hospital days were determined. Segmented regression analyses modeled the association between the time of guideline publication and antimicrobial prescribing; Jan 2010-Aug 2011 and Oct 2011-Jun 2012 were designated as the pre- and post-guideline periods, respectively.  Sep 2011 was excluded from analysis as a transition month. Planned sensitivity analyses assessed the robustness of our estimates to temporal changes in patients' age, chronic conditions, season, study site, length of stay, and ICU admission.

Results: Use of broad-spectrum therapy was stable before the guideline (monthly median 45%, IQR [42, 50]) but declined significantly following guideline publication; a significant increase in narrow-spectrum therapy was also noted. By the end of the study (9 mo. post-guideline), the absolute decline in broad-spectrum therapy use was -15% (95% CI -24, -7) and the absolute increase in narrow-spectrum therapy was 10% (95% CI 2, 20) compared to projected use based on the pre-guideline trend (Figure). Reductions in broad-spectrum therapy use varied by site, ranging from -26% (95% CI -40, -12) to -10% (95% CI -25, 7). Sensitivity analyses were consistent with these findings.

Conclusion: Following guideline publication, use of broad-spectrum therapy declined and narrow-spectrum therapy increased. Changes in prescribing varied by hospital, suggesting that local institutional practices may influence guideline dissemination and implementation.


Derek J. Williams, MD, MPH1, Kathryn Edwards, MD, FIDSA2, Wesley H. Self, MD, MPH3, Yuwei Zhu, MD, MS3, Krow Ampofo, MD4, Andrew Pavia, MD, FIDSA, FSHEA5, Adam L. Hersh, MD, PhD6, Sandra R. Arnold, MD7, Jonathan A. Mccullers, MD8, Lauri A. Hicks, DO9, Anna M. Bramley, MPH9, Seema Jain, MD9 and Carlos G. Grijalva, MD, MPH3, (1)Pediatrics, Vanderbilt University School of Medicine, Nashville, TN, (2)Div of ID, Vanderbilt University Medical Center, Nashville, TN, (3)Vanderbilt University School of Medicine, Nashville, TN, (4)Department of Pediatrics, Division of Pediatric Infectious Diseases, University of Utah Scjpp; Pf

D. J. Williams, None

K. Edwards, None

W. H. Self, None

Y. Zhu, None

K. Ampofo, None

A. Pavia, None

A. L. Hersh, None

S. R. Arnold, None

J. A. Mccullers, None

L. A. Hicks, None

A. M. Bramley, None

S. Jain, None

C. G. Grijalva, None

Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.