Background: C. difficile infection (CDI) is an adverse outcome related to antimicrobial use. Adult antimicrobial stewardship programs (ASP) have demonstrated a decrease in hospital-associated (HA) CDI; however, this effect has not been well explored in pediatric hospitals. The objective of the study is to determine the impact of a pediatric ASP on CDI rates.
Methods: A quasi-experimental study utilizing the Pediatric Health Information Systems (PHIS) between 2004 and 2014 was conducted; the ASP was initiated in March 2008. The ICD9 code 008.45 was used for CDI rates. CDI rates (number/10000 patient-days) and antimicrobial consumption (days of therapy/1000 patient-days) were obtained for each quarter for patients 1-18 years of age. Rates and consumption were compared between the site hospital and peer hospitals. Linear regression was used to determine trends over time. A subgroup analysis was performed with hematology/oncology (H/O) patients.
Results: Antimicrobial consumption decreased for both the site and peer hospitals; however, CDI rates increased over time for both groups (Fig. 1). The CDI rate trend at the site hospital pre- and post-ASP was –0.289 vs. 0.208 per quarter, respectively (difference 0.497, p=0.039). For the peer hospitals, the CDI rate trend was 0.098 vs. 0.211 per quarter, respectively (difference 0.112, p=0.046). In the H/O subgroup: CDI rate trend at the site pre- and post-ASP initiation was 0.465 vs. 0.526 per quarter, respectively (difference 0.061, p=0.923). For the peer hospitals, the CDI rate trend was 0.214 vs. 0.419 per quarter, respectively (difference 0.205, p=0.090).
Conclusion: CDI rates were not impacted by an ASP at a children's hospital. H/O patients are at higher risk for CDI; however, an ASP did not impact the rates in this population. The ICD9 code for CDI does not differentiate between community-associated and HA-CDI and these results may reflect an increase in community-associated CDI which has been demonstrated in previous pediatric studies. More sensitive tracking methods for HA-CDI are needed.
Figure 1: CDI rates and antimicrobial consumption.
J. Newland, Pfizer/Joint Commission: Grant Investigator , Research grant
RPSdiagnostics: Consultant , Consulting fee
J. Goldman, None