Background. Nontyphoidal Salmonella (NTS) is a leading cause of bacterial foodborne illness in the United States. Antimicrobial agents are essential to treat serious illness. CDC's National Antimicrobial Resistance Monitoring System (NARMS) reports annual percentages of antimicrobial resistance among NTS. Incidence estimates of resistant infections have not been reported, in part because data on certain NTS serotypes and resistance are sparse or missing for some states. Using a Bayesian hierarchical model (BHM), a commonly used method to address sparse and missing data, we estimate the incidence of ceftriaxone-resistant NTS infections.

Methods. From 2004 through 2011, 50 states reported NTS isolates received from clinical laboratories to the Laboratory-based Enteric Disease Surveillace (LEDS) and forwarded every 20thisolate to NARMS for antimicrobial susceptibility testing. We defined ceftriaxone resistance as minimum inhibitory concentration ≥4 µg/mL. We used NARMS (number and percent of resistant isolates, total tested), LEDS (number of culture-confirmed infections), and U.S. Census (population) data, aggregated by state and year and grouped into two periods, 2004–2007 and 2008–2011. The BHM incorporated spatially correlated random effects of neighboring states and temporal correlations between periods.

Results. Of 17,498 isolates tested in NARMS, 3.3 % were ceftriaxone-resistant during 2004–2007 and 2.9 % during 2008–2011. Serotypes Typhimurium, Newport, and Heidelberg accounted for 76% of ceftriaxone-resistant isolates. Estimated incidences of resistant infection per 100,000 person-years for 2004–2007 and 2008–2011, with 95% credible intervals (CIs), are: overall NTS 0.41 (0.36–0.46), 0.43 (0.38–0.48); Typhimurium 0.10 (0.08–0.13), 0.11 (0.09–0.14); Newport 0.13 (0.11–0.15), 0.13 (0.11–0.16); Heidelberg 0.05 (0.03–0.06), 0.06 (0.05–0.08).

Conclusion. Using one BHM, we estimated the annual incidence of culture-confirmed NTS infection with resistance to ceftriaxone at 0.41 to 0.43 per 100,000. Estimates for each of the three major serotypes ranged from 0.05 to 0.13 per 100,000 across the study period. This approach has promise for improving incidence estimation in the presence of sparse and missing data, providing information needed to target control measures.

Disclosures.All authors: No reported disclosures.

Author notes

Session: 108. Clinical - Enteric Infections

Friday, October 10, 2014: 12:30 PM

Comments

0 Comments
Submit a comment
You have entered an invalid code
Thank you for submitting a comment on this article. Your comment will be reviewed and published at the journal's discretion. Please check for further notifications by email.