Geographic Distribution of Extended-Spectrum Cephalosporin-Resistant Enterobacteriaceae Across Veterans Health Administration Outpatient Settings

Overall incidence rates of extended-spectrum cephalosporin-resistant (ESCR) Enterobacteriaceae infection at outpatient settings of Veterans Health Administration (VHA) has increased by >100% over 18 years. The researchers performed a retrospective analysis of a cohort of all patients among outpatient settings across the VHA over 18 years who had any positive clinical culture specimen for ESCR collected in an outpatient setting. ESCR was defined by phenotypic nonsusceptibility to ≥1 extended-spectrum cephalosporins or detection of an extended-spectrum beta-lactamase. Patient-level data were grouped by county of residence, and the total number of unique patients who received care within the VHA for each county was used as a denominator. Data were aggregated by time terciles (2000-2005, 2006-2011, and 2012-2017), and overall and county-level incidence rates were calculated as the number of unique patients in each year with ESCR Enterobacteriaceae per person-year.

During the study, there were 1,980,095 positive cultures for Enterobacteriaceae from 870,797 unique patients across outpatient settings of the VHA, for a total of 107,404,504 person-years. Among those, 136,185 cultures (6.9%) from 75,500 unique patients (8.7%) were ESCR. The overall incidence rate was 9.0 cases per 10,000 person-years, which increased from 6.3 per 10,000 person-years in 2000 to 14.6 per 10,000 person-years in 2017. County-level incidence rates ranged widely but increased overall (interquartile range, 2000-2005: 0-6.7; 2006-2011: 0-9.1; 2012-2017: 3.1-14.3 per 10,000 person-years), with some geographic clustering.

This study demonstrates geographic variation, both in incidence rates and trends of ESCR Enterobacteriaceae in outpatient settings of VHA, which suggests the importance of tailoring local antibiotic-prescribing guidelines by incorporating geographic variability in epidemiology.

Source: Goto M, et al. IDWeek 2018. Abstract 1164.

Related Items