BACKGROUND: Effective antimicrobial stewardship requires knowledge of local antibiotic resistance patterns, given geographic variations in antimicrobial susceptibility. Aggregating hospital antibiogram data has been recommended as a feasible, accurate and non-resource intensive method of monitoring community-specific antimicrobial resistance trends. We describe our experience with the practical utility, sustainability, challenges and limitations of such community-level antibiograms over 6 years in Dallas County.
METHODS: From 2009 to 2015, we collected existing cumulative hospital antibiograms from the preceding calendar year from 95% (18/19) of acute care hospitals in Dallas County with facility-specific antibiograms. Aggregated antimicrobial susceptibility data were analyzed and presented according to Clinical and Laboratory Standard Institute guidelines to generate a county-wide antibiogram disseminated to healthcare providers. Trends in proportions of resistant isolates were analyzed using SAS 9.4. Physicians and infection preventionists from participating hospitals were surveyed with standardized questionnaires regarding their perceived usefulness of this community antibiogram.
RESULTS: The 5 most frequently isolated organisms were E. coli, S. aureus. E. faecalis, K. pneumoniae, and P. aeruginosa. Antibiotic susceptibility of the following organisms remained stable in 2014 compared to the prior 5 years: E. coli (n=21,183) to ceftriaxone (92%), meropenem (100%), and ciprofloxacin (71%); K. pneumoniae (n=4,758) to ceftriaxone (92%) and meropenem (99%); S. aureus (n=9,559) to methicillin (52%); S. pneumoniae (n=640) to penicillin (89%); E. faecium (n=929) to vancomycin (34%); and E. cloacae (n=1,237) to meropenem (99%). Susceptibility of P. aeruginosa (n=3,491) to piperacillin-tazobactam declined slightly in 2014 to 82% from the prior 3-year average of 86% (p<0.001). Of the survey respondents from 12 hospitals, all agreed that the County aggregate antibiogram was a valuable tool to monitor antimicrobial resistance trends on a community level, and was helpful to facilitate comparisons with facility-specific antibiograms.
CONCLUSIONS: Our experience supports that an aggregate county-wide antibiogram has been a sustainable and efficient public health tool to conduct surveillance of local trends in antimicrobial resistance among bacteria of public health importance, and to provide awareness of baseline trends to area healthcare providers. Utilization of the community antibiogram to additionally serve as a resource to guide presumptive antimicrobial therapy in healthcare settings without facility-specific antibiograms, or as an impetus to directly prompt changes in practice, however, is currently limited.